starred/windmill
1
0
Fork 0
mirror of https://github.com/windmill-labs/windmill.git synced 2026-10-25 08:03:45 +00:00
Code Issues Packages Projects Releases Wiki Activity
15,334 Commits 844 Branches 1,576 Tags
eda610b2f41ee54ca6983763374dee09bcefd2ee
Commit Graph
102 Commits
Author SHA1 Message Date
AlexRV12 eda610b2f4 [ee] fix: stop handlers, apps and agents resolving untrusted refs (#11893) 
* fix: stop handler, app and agent jobs resolving untrusted references

* perf: skip decoding untrusted args that hold no dollar or escape

* test: drop the transaction the recovery handler no longer takes

* fix: flag reserved names in object keys and tighten the jsonb parity test
 2026-10-21 10:19:20 +02:00
Ruben FiszelClaude Opus 5.5windmill-internal-app[bot]
 		6e565b0dcb feat: route dev workspace and fork jobs to their own worker groups (#11811) 
* feat: route dev workspace and fork jobs to their own worker groups

* fix: harden dev workspace isolation against review findings

* fix: verify hashed dedicated tags and keep dev lineage across renames

* fix: let tag-scoped tokens reach jobs queued on dev workspace tags

* fix: expand tag scopes only in dev workspaces and forks

* fix: expand tag scopes only while dev workspace isolation is on

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* feat: collapse the dev workspace isolation section by default

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* feat: count dev workspace isolation toggles and dev-serving group flags

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: validate the dev workspace tag suffix on every instance config writer

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* chore: update ee-repo-ref to 71e9c0de8317e839f955ebef6a819062f20ac418

This commit updates the EE repository reference after PR #892 was merged in windmill-ee-private.

Previous ee-repo-ref: b5ab84e1705f939c399f6315a03f88b55bce1804

New ee-repo-ref: 71e9c0de8317e839f955ebef6a819062f20ac418

Automated by sync-ee-ref workflow.

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
 2026-10-16 18:47:41 +02:00
hugocasaandClaude Opus 5.5 0b980546b1 feat: agents run on behalf of their deployer (#11430) 
* Revert "docs: agent editor comments describe runs as the caller"

This reverts commit d94919c9fe.

* Revert "refactor: agents run as the caller, with a note on what a shared one needs"

This reverts commit 0748a2fb51.

* fix: a run as an identity continues only the caller's own conversation

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: an agent run as its identity takes only attachments its caller may read

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: the resource editor keeps an agent's identity on an admin's save

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* chore: regenerate workspace tool schemas for preserve_on_behalf_of

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: the rust client sets preserve_on_behalf_of on resource writes

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* feat: agents run on behalf of their identity only when they opt in

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: an agent holds an identity only while it runs on behalf of it

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* chore: reduce the agent identity migration to clearing untrusted keys

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: refuse decoded variable references in inputs to an agent run as its identity

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: user renames and offboarding re-point an agent's identity

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: refuse model-written tool arguments that the tool's job would interpolate

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* chore: agent on-behalf-of migration after main's

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: the agent identity migration also clears trashed agents

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-10-15 19:31:21 +02:00
hugocasaandClaude Opus 5.5 4bc2394322 fix: resolve variables in the url and headers of an mcp resource (#11680) 
* fix(cli): count superadmins as able to preserve ownership on push

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011BzU9zd53JsTShCsbKzYzB

* fix: resolve variables in the url and headers of an mcp resource

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: keep resolved mcp secrets out of header warnings and userinfo urls

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: drop the url from mcp client errors before rmcp logs them

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: run rmcp's reqwest client without a log subscriber

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: scrub the request target from mcp error bodies

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: drop server error bodies from mcp transport errors

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: fail closed when an mcp error message has no status prefix

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
 2026-10-13 13:52:55 +00:00
hugocasaandClaude Opus 5.5 a8e2f1f1ca fix: keep ai agent schema property order stable across jobs (#11687) 
* fix: keep ai agent schema property order stable across jobs

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: keep tool parameter order when filtering preset inputs

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-10-12 11:45:03 +02:00
Ruben FiszelandClaude Opus 5.5 478eef7352 fix: resolve $var: references in mcp resource headers (#11679) 
* fix: resolve $var: references in mcp resource headers

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: require variables:read scope to resolve mcp header variables

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
 2026-10-12 09:56:48 +02:00
hugocasaClaude Opus 5.5Ruben Fiszel
 		d3ddef0b7e feat: retry transient AI provider failures before output reaches readers (#11623) 
* feat: retry transient AI provider failures before output reaches readers

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: keep completed AI streams, stop retrying prompt blocks, bound retry-after

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* refactor: share the AI send-error retry rule between agent and proxy

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: accept a clean end of stream from OpenAI-compatible providers

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: classify AI provider errors the way the reference clients do

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: count a Gemini function call as the end of its turn

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: count an Anthropic stop_reason as the end of its stream

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: accept a Gemini answer that reports usage and no output

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: accept a Gemini stream that closes cleanly with no output

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Ruben Fiszel <ruben@windmill.dev>
 2026-10-09 20:00:50 +02:00
Ruben FiszelandClaude Opus 5.5 cf4bec611f fix: strip NUL from ai agent output and job args before jsonb writes (#11568) 
* fix: strip NUL from ai agent output and job args before jsonb writes

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: also strip NUL from tool call ids and citations

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
 2026-10-06 17:51:07 +02:00
hugocasaClaude Opus 5.5windmill-internal-app[bot]
 		3cba8a7e4c feat: add an AI decision flow step (TypeSafe's Jev, Cloudflare's Clef) (#11474) 
* feat: add an AI decision flow step on TypeSafe's Jev

An `aidecision` step asks TypeSafe's Jev typed questions (choice, score,
yes/no) about a state and returns calibrated answers. It is a plain step:
a flow routes on the answers with a Branch to one after it, which the
editor builds and keeps in step with the questions. It can also be an AI
agent tool, where the calling model supplies the state.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: read single-quoted branch conditions and flag branches on a removed question

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: route decision tools by their tag and keep structured questions as JSON

The questions editor writes only on an edit, so opening a step leaves its
value untouched, and questions the cards cannot hold stay in a JSON editor.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: re-pick the questions editor mode when the value is replaced from elsewhere

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: keep the questions JSON editor's text in step with the value

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix: recreate the questions JSON editor when the value is replaced from elsewhere

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* feat: run AI decisions on Cloudflare's Clef as well as TypeSafe's Jev

A `cloudflare` provider kind reads the hub's `cloudflare` resource (token and
account id), calls the model's Workers AI URL, and unwraps Workers AI's
`result` envelope. Like TypeSafe, it serves decision steps only.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* feat: branch from each AI decision question, and align its editor with the brand guidelines

"Branch on this question" now sits in each choice question's card, which also
links to the Branch to one that routes it and flags options it misses or
branches it no longer matches. Labels, errors, cards and icons follow the form
guidelines and the Branch to one's cards.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* feat: branch on an AI decision from its graph node

A decision with a choice question nothing branches on yet shows a button beside
its node that inserts the Branch to one, as its question card does.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: resolve Cloudflare resources in the AI proxy, and use the design-system button on the graph

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: put the decision_inputs test's doc comment back on it

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* chore: regenerate the flow schema in prompts and the chat after the AiDecision description edit

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* feat: switch AI decision questions between the form and JSON

JSON is always available; the form is offered whenever its cards can hold the
questions, and a value they cannot hold still opens in JSON.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: show step ids as badges and use the JSON toggle of the run forms

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: quote option names in the AI decision branch alerts

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* feat: remove stale branches from the AI decision question card

An empty branch is removed directly; one with steps asks first, naming how many
steps go with it.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* feat: branch on yes/no AI decision questions, and test a decision step from the AI chat

A yes/no question branches like a choice: "Branch on this question" writes one
"yes" branch on `noul >= 0.5`, with the default as "no", and the helper still
reads it once the threshold is tuned. The AI chat's step test runs a decision
alone as a one-step preview, taking the state the chat passes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: record removing several stale branches as one undo step

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: name the typesafe resource type TypeSafe

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: store TypeSafe as the typesafe resource type's display name, not in the word map

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: leave the typesafe and cloudflare resource types to the hub instead of seeding them

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* docs(AGENTS): resource types ship through the hub, not migrations

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: show the questions JSON format beside the JSON toggle

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: give the questions JSON format in the field's description

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: style the decision's graph branch button like Add branch

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: show the questions JSON format only in JSON mode

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* feat: show each question's branching under the questions JSON editor

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: drop the redundant In JSON prefix from the questions format

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: shorten the decision state's description and show an example as its placeholder

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: escape a schema placeholder before the JSON editor renders it as HTML

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* fix: give the decision state's placeholder a text and an object example

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019sEUVZVYBjkzkaVkjbFeAL

* chore: update ee-repo-ref to 154cbccb6b8d47bf662b5507882917b5933d5a45

This commit updates the EE repository reference after PR #839 was merged in windmill-ee-private.

Previous ee-repo-ref: eaf38c1388f3cc5bece163404fffa959ede8cae0

New ee-repo-ref: 154cbccb6b8d47bf662b5507882917b5933d5a45

Automated by sync-ee-ref workflow.

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
 2026-10-05 14:23:34 +00:00
GuilhemandClaude Opus 5.5 35ec16d127 fix: delete a flow chat turn with its flow run, not its step jobs (#11477) 
* fix: delete a flow chat turn with its flow run, not its step jobs

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: give flow-written answers and overlapping turns their own flow run

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
 2026-10-03 20:19:06 +02:00
Ruben FiszelClaude Opus 5.5windmill-internal-app[bot]
 		e7fc1b2e2e feat: restricted job tokens per script and flow (#11484) 
* feat: restricted job tokens (job_token_scopes on scripts and flows)

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: admit flow-run reads, skip dedicated workers, gate on worker version

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: keep restricted jobs off flow runners, preserve scopes on rename and promotion

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: keep restricted jobs off every dedicated handoff, confine progress flow id

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: exclude restricted runnables from dedicated worker startup

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: gate restrictions on the release after 1.821.0

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: store per-job scopes on job_perms instead of v2_job, pin inline runs to the checked version

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* feat: step-level job_token_scopes for flow steps and agent tools

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: fail closed on perms read errors, refuse restricted queue imports, gate step scopes in previews

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: carry a job's scopes on its completion so a re-run keeps the caller's cap

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: carry a zombie job's scopes into its completion

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: leave a zombie for the next sweep when its scopes cannot be read

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* docs: correct the QueuedJobV2 completion comment

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: validate step scopes in batch flows, fail closed on unvalidated step scopes

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: refuse flows with step or tool restrictions at push while an older worker is live

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: apply the step-scope worker gate to flow restarts

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: list the job token toggle with the other step and flow settings

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* chore: pin the EE companion merged with EE main

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* perf: skip scope lookups for unrestricted jobs; list job token setting last

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* style: rustfmt scopes tests

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* fix: confine restricted job tokens to their own run lineage; drop remaining extra lookups

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FHGE3ynoAu6yrgeg4kLwL

* chore: update ee-repo-ref to 259ad3bfeef5285ba80eedc86309b11dca001220

This commit updates the EE repository reference after PR #843 was merged in windmill-ee-private.

Previous ee-repo-ref: 2b77c0225dca441235daf7bf0a06ba968df0c927

New ee-repo-ref: 259ad3bfeef5285ba80eedc86309b11dca001220

Automated by sync-ee-ref workflow.

---------

Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
 2026-10-03 09:33:44 +02:00
hugocasa 039f49ef12 feat: dispatch ai agent tools through the worker queue (#11227) 
* feat: dispatch ai agent tools through the worker queue

* fix: respect worker tags when executing ai agent tools locally

* fix: cancel only direct-child tool jobs on agent failure

* fix: cancel queued descendant tool jobs on agent failure

* fix: route ai agent tools by the flow's tag rule
 2026-09-29 22:30:50 +02:00
hugocasaClaude Opus 4.8windmill-internal-app[bot]Ruben Fiszel
 		c1b59f70dd feat(ai-agent): add compaction memory that summarizes older context (#10928) 
* feat(ai-agent): add autocompacted memory that summarizes older context

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): compact on final-answer turns and count what a turn appended

Address the pre-push review findings on the compaction path:

- A turn the model answers without a tool call left the agent loop on its first
  iteration, so a chat-shaped step never compacted and reloaded the whole
  conversation on every later turn. Compaction now also runs after the loop.
- The trigger measured only the last request, so a single large tool result
  could carry the next one past the window without ever crossing 80%.
- The summarization call re-sent the usage-tracking request shape on endpoints
  the loop had already learned to drop it for.
- The flat 8000-token summary reserve swallowed the whole target on a small
  context window, leaving one message in the tail and summarizing the rest.
- A response cut off inside the <analysis> scratchpad was accepted as a summary.
- The chat-mode memory default was a shared object the step form edited in place.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): keep Anthropic prompt counts and compact once per response

Address the first CI review round on the compaction path:

- Anthropic's streaming parser dropped `message_start`, the only event carrying
  the prompt-side counts, so a native Anthropic run reported no input tokens at
  all and compaction fell back to a character estimate.
- A loop that exits without issuing another request — a structured-output turn
  does — reached the post-loop pass still holding the previous measurement and
  compacted a second time, or retried a failure with nothing changed.
- The summarization call inherited the step's `max_completion_tokens`; a low one
  truncates the summary inside its scratchpad, which counts as a failure and
  disables compaction after three of them.
- A fired trigger that found nothing to summarize said nothing.
- Memory already over the window — a lowered `context_window`, or a step moved
  over from `auto` — had no way back, since compaction only ran after an
  accepted request. It now also runs once before the first one.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): state the summary's own completion cap and drop the pre-flight pass

- The summarization call asked for no completion cap at all, which is "uncapped"
  only on the OpenAI-shaped providers: Anthropic substitutes 64000, over several
  Claude models' output ceiling, and Bedrock leaves the model's own small default,
  short enough to cut the response off inside its scratchpad. It now asks for the
  reserve the split already set aside, raised to the step's cap when that is larger.
- Compaction no longer runs before the first request. The fallbacks the loop learns
  from a rejection are not known that early, so on exactly the endpoints that need
  them the summarization was malformed by construction: it failed, spent a strike,
  and the first agent request still carried the oversized conversation. A memory
  already past the window is repaired on the turn after a request the endpoint
  accepts, rather than by a pass that cannot succeed there.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): ask the summary for exactly the room the split reserved

The split scales its reserve down on a small window while the request asked for
a flat 8000, so the two diverged below an 80k window: on a 4k/8k model the cap
alone exceeded the window and every summarization was refused, and on a 20k one
a full-length summary could land the conversation back over the trigger and
compact its own previous summary on the next response. Both now read one
`summary_reserve_tokens`.

The call also no longer inherits the step's reasoning effort. Every provider
counts thinking against that same budget, so a high-effort model could spend the
whole reserve before writing anything and return a summary cut off inside its
scratchpad; the compaction prompt asks for an `<analysis>` block, which is the
reasoning this call needs.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): charge the compaction budget for tools and the system prompt

The tail budget was the whole target, but a request also carries the system
prompt compaction keeps and the tool definitions, which are not in the message
list at all. On a small window those are most of it: a tail sized to the full
target left the next request back over the trigger, compacting again every
response, and the no-usage estimate missed the tool schemas entirely so it could
fail to trigger at all. Both now account for them.

The reserve also gains a floor. It is the summary's output cap as well as the
room the split leaves, and scaled down without one a small window gave a
structured nine-section summary a few hundred tokens — truncated inside its
scratchpad every time, which is discarded, which switches the mode off after
three.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): count Gemini's tool-use prompt tokens in an agent step's usage

Gemini splits a tool-using turn's input across `promptTokenCount` and a disjoint
`toolUsePromptTokenCount`, and its thinking apart from `candidatesTokenCount`.
The agent step's parser read only the headline fields, so every tool-using turn
under-reported both — and the compaction trigger, which runs off the reported
prompt, could not see the tool results that grew it. It now goes through the same
helpers the proxy path already used.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): calibrate the compaction estimate against the measured prompt

Two rounds running, the finding was "the character estimate cannot see input X"
— tool schemas, then S3 attachments, which are short paths in the message list
and whole images by the time a provider counts them. Enumerating those is a list
that only grows, so the estimate is now scaled to the one number that is ground
truth: what the provider charged for the last request. Attachments, tokenizer
drift and whatever comes next fall out of that, because the estimate is only
ever used relative to itself.

Also stop the Gemini helpers turning an absent count into `Some(0)`. Downstream,
absent means "fall back to estimating the conversation" while zero reads as an
empty prompt and would hold the trigger below its threshold for the whole run.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): charge attachments what they cost and let a heavy short prefix compact

The calibration conserved the conversation's total cost but spread it by
character count, so an attachment — a short S3 path in the message list, a whole
image or PDF once a provider expands it — was charged to the text messages around
it and stayed nearly free in the split. It now carries a nominal cost of its own,
which the calibration corrects a residual on rather than the whole gap.

The four-message minimum also refused exactly the case that fix is for: an
attachment arriving on the first or second turn can pass the trigger before four
removable messages exist, and summarizing even one of them saves most of the
prompt. A prefix worth a quarter of the window is now enough on its own.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): never summarize a prefix holding only a previous summary

The message-count floor was carrying a second job: a fresh summary sits in a one
or two message prefix, so requiring four declined it. The share threshold added
last commit admits it, and a summary is reserve-sized by construction — so the
post-compaction shape could spend one summarization per response swapping a
summary for another the same size, shrinking nothing and losing fidelity each
time. A previous summary no longer counts towards that threshold.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): take the context window from the model and drop the estimate calibration

Brings compaction in line with how the AI session does the same job, which had
already answered these three questions.

- The window is looked up from the model. `MODEL_CONTEXT_WINDOWS` in
  `windmill-ai/src/model_context.rs` mirrors the session's table in
  `copilot/modelConfig.ts`, entry for entry and with the same matching rules;
  each side points at the other, since a model added to one and not the other
  compacts at two different sizes. A step's `context_window` becomes the
  override for what the lookup cannot serve, and chat mode writes none.
- Provider usage is normalized where the provider's quirk is, not at the
  consumer. `TokenUsage::with_cache_beside_input` raises `input_tokens` to the
  whole prompt for Anthropic and Bedrock, which report their cached prefix
  beside it; the OpenAI shape already counts it inside. `prompt_tokens()` is
  then just `input_tokens`, rather than inferring the shape from whether a
  write count is present.
- The estimator is no longer calibrated against the measured prompt. The
  session uses the provider's count when it has one and a chars/4 estimate
  otherwise, with nothing in between, and a tail sized a little wrong only
  compacts again a turn later.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* feat(ai-agent): summarize memory down to what the database can store

Without an instance object store, memory is a 100KB database row cut
from its oldest message, the summary included, so compaction on a
mainstream model never got to keep anything across runs. A step that
persists there now runs its post-loop compaction pass against the
smaller of the model's window and the cap at chars/4, about 25k tokens:
the loop keeps the whole window, and what is written is a summary plus a
tail that fits. The run logs when that pass summarizes, and how many
messages the write dropped when one still overshoots.

The editor's storage warning on the option is removed: nothing exposes
the instance storage to it, so it keyed on the workspace S3 setting,
which is unrelated to where memory goes.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): get a complete, billed summary out of every provider

Compaction against the real providers turned up four things the stub
could not: Gemini and OpenAI's reasoning models think by default and
bill it against the same cap the summary must fit in, so the
summarization request now asks them for their least (none, low); an
OpenAI Responses call that hits max_output_tokens ends in
response.incomplete, whose usage the parser dropped, so that
summarization went unbilled; a summary that quotes </summary> when it
describes its own instruction was cut off at the quote, on the agent
step and the AI session alike; and the prefix could end on an unanswered
user message, after which the instruction reads as part of that turn
(Anthropic merges the two outright). The tail now starts on a user
message, and both prompts tell the model the instruction is not part of
the conversation.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): compact down to half the window, on the agent step and the AI session

The gap between the 80% trigger and the target is what one compaction
buys, and every summarization request carries most of the window. At a
70% target a 128k model summarized about 13k tokens of prefix for a
summary of up to 8k, so each ~100k-token request bought a few turns of
room before the next one re-summarized the previous summary. At 50% the
same request frees about 30k.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): drop the workspace-S3 memory hint and state the database bound in the tooltip

The memory field warned that memory is kept in the database whenever the
workspace had no S3 storage. That setting has no bearing on where memory
goes: the instance object store decides, and nothing exposes it to the
editor. The field's tooltip now describes both memory kinds and states
the database bound unconditionally; the run log says what happened.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): send the summarizer its tool history as text

The summarization request carries no tool definitions, and Bedrock's
Converse API rejects toolUse/toolResult blocks that arrive without them,
so on Bedrock every summarization of a prefix holding a tool call failed
silently until the breaker tripped. The prefix's tool calls and results
now reach the summarizer rendered as text, on the agent step and in the
AI session's compaction, which goes through the same proxy.

Also drops the TokenUsage::prompt_tokens accessor, which had become a
plain read of the normalized input_tokens, and shortens the context
window field's description.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): price attachments from the provider count, bound storage in bytes, effort per pro model

Addresses two Codex rounds and a leftovers audit.

- Attachments were priced at a flat 1500 tokens in the split, so a
  multi-page PDF (tens of thousands of tokens to the provider, a short
  S3 path in the message list) could be kept in the tail or leave no
  prefix worth summarizing. They are now priced from the provider's
  count for the request that carried them, less that request's text,
  with the 1500 floor where nothing was counted.
- The database storage bound measured the provider's token count, but
  the 100KB cap is bytes and repetitive text packs several characters
  per token. The persist pass now measures the serialized conversation.
- The summarizer forced `low` on every reasoning model, which the pro
  variants reject (gpt-5-pro takes only high, gpt-5.2-pro starts at
  medium); they now get no effort.
- Dropped the unused prompt_tokens accessor and its orphaned assert, an
  unused PartialEq, a needlessly public lookup, and fully-qualified
  Gemini calls; refreshed stale comments and the memory_id schema doc;
  regenerated the flow schema artifacts.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): evict a heavy attachment into the summarized prefix, not the tail

Pricing attachments from the provider count was not enough on its own: a
leading attachment is a user message, and the boundary rule pulled the
last unanswered user turn back into the kept tail to keep it with its
answer. For a heavy attachment that dragged it into the tail — or, at
the front, emptied the prefix — so it was never summarized and rode
every request. The boundary now moves forward instead, keeping that
user turn and its answer in the summarized prefix. Verified on the
running instance: a 25k-token PDF on a 30k window is summarized out on
the turn it overflows, and later turns drop from 26k to ~1.5k tokens.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): keep the forward boundary move off tool results and the prefix start

The forward move that keeps an unanswered user turn out of the tail had
two edges the third Codex round found: advancing past the user could
land the boundary on a tool result (its tool_calls then summarized away,
orphaning it), and with no system prompt the summarizable prefix starts
at 0, so a trigger firing while the tail estimate fit everything indexed
below the start and panicked the task. The forward scan now skips
tool-opening boundaries, and the move is guarded above the prefix start.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): drop the step temperature from the summary request

OpenAI's reasoning models (gpt-5-mini, gpt-5.1, gpt-5.2) reject
`temperature` alongside any reasoning effort but their own default, so a
step configured with a temperature made every summarization fail once
the summarizer forced a low effort — history then grew unchecked. The
internal summary call now omits the step's temperature: a structured
extraction does not need a set one, and omitting it sidesteps each
provider's temperature-versus-reasoning rules. Confirmed against the API
that low + temperature is refused on those models.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): compact an oversized loaded memory before the first request

Compaction was reactive, taken only after a request the endpoint
accepted, so the fallbacks the loop learns from a rejection are known
first. But a memory loaded from an earlier run can already exceed this
run's window — the step was switched to a smaller model, or a run under
a wider one persisted more than fits — and that first request then
overflows and fails the run, with every retry reloading the same
history and failing again. A pass is now taken up front, off the
character estimate, before the first request. It uses the default
request shape; an endpoint needing a fallback may reject this one
summary, which is non-fatal, and mainstream providers need none.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): under the storage bound, trigger on the max of bytes and model tokens

The storage-bound pass measured only the serialized row size, so an
attachment — a few bytes as an S3 path but nearly the whole model
context — read as tiny and the pass skipped a compaction the model
needed. It now takes the larger of the byte measure and the model's
token count, since repetitive text is few tokens but many bytes and an
attachment is the reverse; either being over must fire a pass.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): drop oldest turns when a summary cannot fit the window, as the AI session does

An oversized loaded memory (a step switched to a smaller model, or an
object-store run that persisted more than a later model's window holds)
left a prefix larger than the summarizer's own window, so the summary
request overflowed and failed, the memory was untouched, and every
retry failed the same way. The AI session handles this by falling back
from summarization to dropping the oldest turns down to the target;
compaction here now does the same. When a summary cannot run — it
failed, the breaker is tripped, or nothing is worth folding — the oldest
turns are dropped until the conversation fits and opens on a user
message, keeping the newest turn. The next request then always fits.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): drop whole turns only, keep the storage pass to bytes, refresh the count after a rewrite

Three edges the seventh Codex round found, all in the drop-oldest
fallback and the storage-bound measure:

- drop_oldest_to_fit dropped to any point that freed enough, which could
  strand a tool result whose tool_calls went with the messages before
  it. It now drops whole turns only, always landing the boundary on a
  user message and never splitting the newest turn; a lone turn too big
  for the window is left whole rather than broken.
- The storage-bound pass measured the whole model prompt against the
  shrunk 25k window, so a large tool roster and the system prompt —
  neither written to the row — tripped it on a conversation the row
  easily held. It measures the serialized bytes alone now; the model's
  own window is enforced by the in-loop passes and the pre-first-request
  pass, so the persisted size is all this pass is for.
- A compaction rewrites the message list, so the provider's count for
  the request that produced it no longer lines up. The count is now
  cleared after any pass that rewrites the conversation, so a later pass
  measures the estimate over the actual messages instead of a stale,
  larger prompt (which could decline a summary that already fit and then
  drop it). The step temperature, no longer sent to the summarizer on
  any path, is dropped from the request struct.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): measure only the persisted messages against the storage cap

Persistence strips the system prompt before writing the memory row, but
the storage pass was serializing every message including it, so a large
system prompt with a tiny conversation reported far over the storage
trigger, and the fallback dropped the one real turn, run after run. The
storage measure now serializes only the non-system messages, matching
what the row actually holds.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix(ai-agent): run the model-window pass before the storage-bytes pass post-loop

A turn the model answered without a tool call broke before the in-loop
compaction check, so on database-backed memory its only pass was the
storage one, which measures bytes. An attachment fills the model context
but is a few bytes in the row, so that turn never compacted and a
follow-up could overflow the model. The post-loop now runs a
model-window pass first, off the provider's count, then the
storage-bytes pass when the row is smaller than the model — both limits
enforced for a chat-shaped step, not just the one that happens to bind.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix: simplify agent compaction and preserve execution history

* fix: remove unused compaction history setting

* fix: preserve answers and recover rejected agent context

* refactor: make agent compaction transactional

* fix: skip agent summaries that cannot fit retained context

* fix: explain skipped agent context compaction

* fix: retain recent agent memory when storage compaction cannot fit

* fix: start retained agent memory at a user turn

* fix: reject unsafe agent memory truncation on storage fallback

* docs: clarify agent context window override scope

* fix: keep recent turns verbatim when compaction memory outgrows storage

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix: keep the compaction summary out of the agent's answers

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* fix: shorten the agent context window help text

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViJyjUmidDYV2m6ifQdLeH

* chore: update ee-repo-ref to 942d4013f36edac1fc9a9addbdb02198db1c7a05

This commit updates the EE repository reference after PR #812 was merged in windmill-ee-private.

Previous ee-repo-ref: 8ca1682ce6106ba6ea96894fbe606dac64102eb6

New ee-repo-ref: 942d4013f36edac1fc9a9addbdb02198db1c7a05

Automated by sync-ee-ref workflow.

---------

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
Co-authored-by: Ruben Fiszel <ruben@windmill.dev>
 2026-09-29 19:01:53 +02:00
Ruben FiszelandClaude Opus 5.5 649c43e7c1 fix: run an AI agent tool on the worker its own tag selects (#11370) 
* fix: run an AI agent tool on the worker its own tag selects

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: run a tagged agent tool inline when this worker serves its tag

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: give an inline agent tool a job token of its own

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

* fix: report a lost tool wait to the model and cancel tools on agent timeout

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
 2026-09-26 17:32:25 +00:00
GuilhemandClaude Opus 5 a571117f3f feat: store mcp tool call, result and reasoning on flow conversation rows (#11176) 
* feat: store mcp tool call, result and reasoning on flow conversation rows

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: keep the failure reason and web search citations on tool rows

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: persist a structured answer as its own row and keep reasoning-only rows from closing a turn

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: keep a failed Windmill tool's error on its conversation row

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: let a structured answer row claim its streamed thinking

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: scope a structured answer's claim to its own turn and document the row fields as stored

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: claim a textless assistant row only within the newest turn

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* refactor: store the thinking that led to a tool call on the tool row

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix: keep tool calls in stream order and scope a structured answer to its turn

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* feat: store the files a user message carried as object-storage references on its row

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* feat(chat-sdk): read a message's attachments and build their download URL

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* feat(chat-sdk): carry a loaded user message's attachments in AI SDK metadata

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* feat: store the model's call and what it got back on every tool row

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix: pass no extras in the orphaned-conversation test's message insert

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(chat-sdk): keep a stored JSON null tool result instead of the row text

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix: label a tool call the turn finished without as not finished

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix: keep a structured answer's streamed call out of the did-not-finish label

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix: write a turn's conversation rows in order and describe stored tool calls

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix: await the image answer row like the agent loop's other rows

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(chat-sdk): place a row nothing streamed by its sequence, not at the end

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(chat-sdk): place only tool rows by sequence, keep a closing row last

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* fix(flow chat): show a failed tool's error, and no Retry on a running or stopped turn

A failed tool card showed the row label in place of the error the row now
stores. The live tool result now reports failures, so a failed call briefly
flagged a running turn as failed; a stopped turn, whose last row is the
failed tool or the cancelled flow's failure, offered Retry too.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-17 15:35:18 +02:00
hugocasaandClaude Opus 5 c297ed0052 feat: managed memory with an inherited or custom memory id per step (#11118) 
* feat: split ai agent memory into agent policy, run memory id and step history

* fix: scope string memory ids to workspace and flow, keep nested tool history inputs

* chore: update sqlx cache for the flow context query

* docs: describe memory id scoping as collision-free rather than isolated

* chore: regenerate openflow json after merging main

* fix: offer no memory id for legacy manual memory, document linked history inputs

* fix: seed provided messages from legacy manual memory and hide its note once set

* fix: bypass memory when a provided messages expression evaluates to null

* fix: require a user message when provided messages are empty

* chore: keep the empty messages comment within the line width

* docs: name the history inputs wherever linked steps list their flow-local inputs

* docs: keep the memory storage path on one line

* feat: managed memory with an inherited or custom memory id per step

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix: list a custom memory id in the test run form and name where an inherited one comes from

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix: keep memory id out of the add-field menu and drop the memory id telemetry

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix: keep legacy auto memory without an id working after an untouched redeploy

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix: rename step messages to previous_messages and address review

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* style: rewrap comments and docs lines lengthened by the previous_messages rename

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* refactor: read agent memory as either a legacy shape or the current one

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix: name the memory setting in ignored-input notes and keep conversions honest

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix: keep a legacy memory count unset on open and read a cleared count as off

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: address review on cleared test history and zero-count memory

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: drop flow-local keys from a linked agent resource before interpolating it

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: keep a linked resource's own inputs as fallbacks and note ignored history on image runs

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: restore the linked agent draft tests and log ignored history on every image run

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: resolve the one-of variant from the value when the selected one leaves the list

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: treat zero-count managed memory as off when enabling chat mode and shorten comments

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix: stop requiring user_message in the openflow agent contract when previous messages are the prompt

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
 2026-09-17 11:55:43 +02:00
Ruben Fiszel 2cb02e3b33 fix: offload php signature parsing from async workers (#11027) 
* fix: reduce php parser stack use in debug workers

* test: document php stack regression expression depth

* fix: offload php signature parsing from async workers

* fix: address php parser review nits
 2026-09-08 15:20:02 +02:00
Ruben Fiszel 02c4a9e515 fix: carry the token label into job-run audit rows (#10433) 
* fix: carry the token label into job-run audit rows

* docs: state the audit end-user precedence at the push signature

* chore: point ee-repo-ref at the companion branch

* docs: state the username/end_user split at the push signature

* feat: keep the audit caller searchable when a token label takes end_user

* fix: skip the caller parameter when it repeats the end user
 2026-07-31 16:13:12 +02:00
Ruben Fiszel 7d097d25c3 fix(ai): pass only the output of a nested agent tool to the parent (#10416) 2026-07-30 15:04:05 +00:00
hugocasa 87f8d46aaf fix(ai-agent): align agent_actions_success with agent_actions for mcp and websearch (#9983) 2026-07-07 16:19:37 +02:00
Ruben FiszelandClaude Opus 4.8 317a8629d1 fix(ai): enforce resource authz when loading MCP tools in agent worker (#9571) 
* fix(ai): enforce resource authz when loading MCP tools in agent worker

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* test: describe MCP authz test invariants instead of drafting history

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-06-14 23:50:18 +02:00
Ruben Fiszel a6a5600833 feat(flow): support worker tag override on AI agent steps (#9513) 
* feat(flow): support worker tag override on AI agent steps

* refactor: drop ineffective tag passthrough in nested agent tool path

* chore: regenerate openflow-derived system prompt artifacts
 2026-06-10 13:54:05 +00:00
Ruben FiszelandClaude Opus 4.8 8053266f88 fix(mcp): resolve MCP resource token via caller RLS + SSRF-guard url (#9428) 
* fix(mcp): resolve MCP resource token via caller RLS + SSRF-guard url

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* fix(mcp): clone user_db for oauth2 refresh and drop advisory ids from comments

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* fix(mcp): disable redirects on MCP client to prevent SSRF bypass

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-06-03 09:00:43 +00:00
centdix a7d85a39ff refactor: clean up ai provider proxy logic (#9360) 
* refactor: clean up ai provider proxy logic

* docs: remove completed ai refactor plan

* fix: audit failed google global proxy calls
 2026-05-28 16:05:04 +00:00
centdix 17cf538a2d refactor: move ai providers to windmill-ai (#9120) 2026-05-12 14:09:31 +00:00
centdix 6f7d31e56b refactor: move ai image handling to windmill-ai (#9098) 2026-05-11 14:59:40 +00:00
centdixandwindmill-internal-app[bot] 27acbbf3d5 refactor: move ai sse plumbing to windmill-ai (#9059) 
* docs: refine windmill ai refactor plan

* refactor: move ai sse plumbing to windmill-ai

* refactor: remove ai re-export shims

* fix: update ee ai memory ref

* chore: update ee-repo-ref to d3bc7fa85195b46b7a38d43c2f806520bf8b5454

This commit updates the EE repository reference after PR #560 was merged in windmill-ee-private.

Previous ee-repo-ref: ff35bf7cc198e13884b33654e1d6dbd8a8b314d3

New ee-repo-ref: d3bc7fa85195b46b7a38d43c2f806520bf8b5454

Automated by sync-ee-ref workflow.

---------

Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
 2026-05-11 10:01:45 +00:00
hugocasaClaude Opus 4.7Ruben Fiszel
 		70a5880d36 fix: distinguish AlreadyCompleted from execution failure on OTLP job span (#9004) 
* fix: distinguish job-already-completed from execution failure on OTLP span

handle_queued_job's bool return type conflated two distinct Ok(false)
cases: a real race with another worker (Error::AlreadyCompleted) and any
job execution that returned an error via process_result. The outer "job"
span recorded "job already completed by another worker" for both, so
every failed Python/bun/etc. script ended up with that misleading
otel.status_message even though the real error was correctly recorded
on the inner job_postprocessing span.

Replace the bool with a JobOutcome enum (Completed / Failed { description }
/ AlreadyCompleted). Failed carries the truncated error string, so the
outer span's Status.message now reflects the actual cause.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix: extract user-facing error from error_value for span description

Cubic flagged that capturing e.to_string() before the match meant
ExitStatus failures (the most common failure mode for script jobs)
ended up with the generic "exit status: …" string rather than the
script error extracted from job logs by extract_error_value.

Move the description capture to after error_value is built and
deserialise it as ErrorMessage to pull out the structured message.
Falls back to "Job failed" if the value isn't shaped that way.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix: use permissive description extraction for OTLP status message

Cubic flagged that strict ErrorMessage parsing downgraded real failures
to a generic "Job failed" whenever the result wasn't shaped as
{message, name} — e.g. agent-worker's "See logs for more details" raw
string, or runtime-written result.json files with a different shape.

Switch to parsing as serde_json::Value and pulling .message out if it's
a string, falling back to the whole value if it's a bare string. Only
fall back to "Job failed" when neither is available.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-authored-by: Ruben Fiszel <ruben@windmill.dev>
 2026-05-01 18:25:18 +00:00
Ruben FiszelandClaude Opus 4.7 aedf369174 fix: pair PG arg type with actual Rust binding to keep query_typed_raw safe (#8999) 
* fix: pair PG arg type with actual Rust binding to keep query_typed_raw safe

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix(pg): wrap encoder errors with arg context, add fallback test

Followups on #8999 review:

- Wrap rust-postgres "error serializing parameter N" failures with the arg
  name, JSON value kind, and asserted Postgres type plus a hint about an
  explicit cast — so users see actionable context instead of an opaque
  WrongType.
- Drift-prevention meta-test: assert otyp_to_pg_type and convert_val agree
  on the Type for every recognised arg_t when the JSON value matches its
  natural Rust kind. Catches future drift if either side changes.
- Integration test for the prepare + query_raw fallback path: confirms
  unrecognised arg_t (custom enum) is routed through prepare and the
  server-resolved type appears in the failure surface — flips into a
  test failure if a regression accidentally routes unrecognised types
  through query_typed_raw.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(pg): add otyp_inferred flag + regex-based placeholder renumbering

Two follow-ups from the review of #8999:

1. **Issue #1 (Number/Bool + explicit text decl in WHERE)**

   Add `Arg::otyp_inferred: bool` to the parser. The PG SQL parser sets
   it `true` only at the "no info → fall back to text" site (bare `$N`,
   no inline cast, no `-- $N (TYPE)` decl). All other arg sources keep
   it `false`.

   In `convert_val` this flag distinguishes:
   - explicit text-like target (`-- $1 (text)` or `$1::text`) — coerce
     `Bool`/`Number` → `Box<String>` so `WHERE text_col = $1` works
     (`text = text` operator). Pre-#8988 behaviour, restored.
   - parser-default text (bare `$N`) — bind the value's natural Rust
     type so the regression case (`Value::Bool` against a real `bool`
     column via `CAST AS bool`) keeps working.

   `Arg` is in `windmill-parser`; the new field has `#[serde(default)]`
   so persisted signatures stay backward-compatible.

2. **Issue #4 ($5/$50 substring rewrite collision)**

   Replace the per-index `String::replace` chain (which turned `$50`
   into `$10` when oidx=5 was processed first) with a single regex
   pass. `\d+` is greedy, so `$5` and `$50` match as distinct units;
   indices outside the mapping are left intact.

3. Tests:
   - parser: `test_parse_pgsql_otyp_inferred_flag` covers bare/inline-
     cast/decl/mixed shapes.
   - executor unit: `convert_val_bool_against_every_arg_t` and
     `convert_val_*_number_*` split each text-like target into explicit
     vs inferred expectations.
   - executor unit: `renumber_sparse_placeholders_no_collision`.
   - integration: `test_postgresql_arg_type_combinations` adds 4 cases
     covering decl(text)+Number/Bool in WHERE, bare $1+Bool, and
     sparse positional args ($5/$50).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(pg+sdk): enum support, extended String arms, position-aware $N rewrite, SDK quality

Backend:

1. **`AnyTextValue` ToSql/FromSql wrapper**: vanilla `tokio_postgres`'s
   `ToSql for String` / `FromSql for String` reject `Kind::Enum` and
   `Kind::Domain` even though the wire format is plain UTF-8. The wrapper
   accepts those kinds in both directions. End result: explicit
   `$1::my_enum` / `CAST($1 AS my_enum)` casts now round-trip without the
   ugly `CAST($1::text AS my_enum)` workaround, AND `SELECT enum_col`
   results come back as JSON strings instead of erroring at the FromSql
   layer.

2. **#10 — Value::String → numeric/real/double/oid/bool**. Without these
   arms, a string-encoded value (`"3.14"`, `"true"`) for a non-text /
   non-temporal arg_t fell through to `Box<String> + TEXT`, which then
   failed at the server (no implicit cast text→numeric in expression
   context). Now strings are parsed into the matching native type with
   clear error messages on parse failure.

3. **Position-aware `$N` rewrite**: replaces the regex-based renumbering
   (which fixed the `$5/$50` substring collision but still walked through
   string literals and comments, mangling `'price: $5'` etc.) with a
   walk over `parse_pg_statement_arg_positions` — the same
   string/comment/dollar-quote-aware tokenizer used for index discovery.
   Adds `parse_pg_statement_arg_positions` to the parser's public API.

SDK:

4. **BigInt support**: `JSON.stringify(BigInt)` throws. The SDK now
   stringifies bigints before serialisation; the executor accepts
   numeric strings into BIGINT arg slots via the existing
   `Value::String → INT8` parsing arm. SDK-side `inferSqlType` is split
   so `BigInt` always resolves to `BIGINT` (was reaching
   `Number.isInteger(BigInt)` which returns false → wrong default).

5. **Homogeneous array auto-tag**: `${[1,2,3]}` against an `int[]` column
   now emits `$1::BIGINT[]` instead of `$1::JSON`. Detection covers
   primitive types only (number / bigint / string / boolean); mixed or
   nested arrays still fall back to JSON. Mixed int/float widens to
   `DOUBLE PRECISION[]`.

6. **`.query()` positional bug**: previously the `.query()` method
   abused the template-tag builder, which appended `$N::TYPE` after the
   user's literal SQL string instead of binding by position
   (`SELECT $1, $2` became `SELECT $1, $2$1::BIGINT`). Now `.query()`
   builds the executor-shaped content directly: a `-- $N argN (TYPE)`
   declaration block followed by the user's SQL verbatim.

Tests:

- Parser: `test_parse_pg_statement_arg_positions_skips_strings_and_comments`
  asserts string literals, comments, and dollar-quoted blocks don't
  produce positions (so renumbering doesn't mangle them).
- Executor unit: `renumber_sparse_placeholders_no_collision_no_string_mangling`
  uses the new position-aware path and includes string-literal + comment
  + `$$…$$` cases. Existing convert_val tests grow to cover new
  String→numeric/real/double/oid/bool arms.
- Integration: `test_postgresql_arg_type_combinations` adds 13 cases
  (enum round-trip both directions, string→numeric/real/double/bool/oid,
  string-literal `$N` non-mangling). The prepare-fallback test now
  asserts SUCCESS (not failure) for enum encoding via AnyTextValue.
- SDK: new `typescript-client/tests/sqlUtils.test.ts` (42 tests)
  exhaustively covering inferSqlType primitives + arrays,
  parseTypeAnnotation, datatable() template tag (with all the new
  shapes — BigInt, homogeneous arrays, RawSql, schema preamble),
  datatable().query() positional, and ducklake() shape.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(pg): replace DISCARD ALL with curated reset (preserves typeinfo cache)

Found while exhaustively probing custom-type DX: every cached-connection
reuse was running `DISCARD ALL`, whose included `DEALLOCATE ALL`
deallocates *all* prepared statements server-side — including the typeinfo
statements that tokio_postgres caches per-Client to resolve custom enum /
domain Oids. tokio_postgres still held `Statement` objects whose names
the server had forgotten, so the next custom-type query failed with
intermittent "prepared statement \"sN\" does not exist" errors. The
failure was easy to reproduce: any sequence that forced typeinfo lookup
for two different custom-type kinds on the same cached connection (e.g.
enum followed by domain) would hit it.

Replace `DISCARD ALL` with a curated reset that explicitly targets the
state we actually care about, *without* touching prepared statements:

  RESET ALL                     — GUC parameters (search_path, application
                                  _name, statement_timeout, …)
  RESET SESSION AUTHORIZATION   — undoes both `SET SESSION AUTHORIZATION`
                                  and `SET ROLE` (RESET ALL does NOT —
                                  these aren't GUC parameters, so without
                                  this an elevated role from a previous
                                  job would silently leak)
  UNLISTEN *                    — drops LISTEN registrations
  CLOSE ALL                     — closes open cursors

Trade-off: temp tables, advisory locks (session-scoped), and user-created
PREPARE statements may persist across cached-connection reuse — rare in
datatable / PG-script workloads. tokio_postgres's typeinfo cache survives
intact, so custom enum / domain queries are fast on subsequent reuse.

Tests:
- `test_postgresql_custom_types_on_cached_connection` — runs 10×
  alternating enum + domain queries on a cached connection. Pre-fix this
  failed with `prepared statement "sN" does not exist` after the first
  reuse; post-fix passes.
- `test_postgresql_set_role_does_not_leak_across_cached_connection` —
  switches `SET ROLE` and `SET SESSION AUTHORIZATION` to a non-postgres
  role, then runs a follow-up job and asserts current_user/session_user
  are restored. Specifically catches the case where someone might switch
  back to `RESET ALL` alone (which doesn't cover SET ROLE / SESSION
  AUTHORIZATION) and silently introduce a permission-leak vector.
- All existing session-isolation tests
  (`test_postgresql_cached_connection_resets_session`,
   `test_postgresql_single_worker_session_isolation`,
   `test_postgresql_100_jobs_cached`) continue to pass.

Found via end-to-end probing of datatable / PG-script DX, not previously
covered: the existing isolation tests only did `SET ROLE postgres`, the
connecting user, so the leak was invisible.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(pg): address PR #8999 review (cubic + claude)

cubic (P1, real bug):
- `convert_vec_val` for `timetz` array asserted `Type::TIMETZ_ARRAY`, but
  chrono `NaiveTime` only encodes for TIME (same caveat as the scalar
  arm). Switch to `Type::TIME_ARRAY`; rely on PG's implicit `time→timetz`
  assignment cast at the column site. Add an explicit unit test.

claude (#1, silent failure → explicit error):
- `Bool` + explicit `(char)` / `(character)` decl previously silently
  bound BOOL, hoping the server would cast at the use site — but PG has
  no implicit `bool→char` and the resulting error
  ("operator does not exist: bool = char") was opaque. Now error at
  bind time with an actionable hint to use `bool` decl or pass the
  value as a "t"/"f" string.

claude (#2, asymmetry doc):
- Object/Array still coerce to text on `matches!(typ, Typ::Str(_))`
  (covers both explicit AND inferred-default text), unlike Bool/Number
  which key on `explicit_text_target`. The asymmetry is intentional
  (no implicit `jsonb → text` cast in expression context vs PG having
  implicit `bool/int → text` casts) — added a body comment so future
  maintainers don't try to "align" them.

claude (#3, perf):
- `parse_pg_statement_arg_indices` and `parse_pg_statement_arg_positions`
  walked the SQL tokenizer twice. Fold into a single pass that derives
  the index set from the position list.

claude (#4, fmt drift):
- `cargo fmt` over the parser crates I touched with perl scripts in the
  earlier commit (windmill-parser-{sql,bash,ts,go,php,java,csharp,nu,py,
  rust,graphql,yaml,r}). Net cosmetic.

claude (#5, parseTypeAnnotation):
- One-line caveat in the SDK's `parseTypeAnnotation` that the returned
  string is presence-only (e.g. `${x}::DOUBLE PRECISION` returns
  `"DOUBLE"`, `CAST(${x} AS int)` returns `"int)"` — neither matches a
  real PG type, but the only consumer just checks `!== undefined`).

While here — discovered + fixed independently while exhaustively probing
DX:

- **Replace `DISCARD ALL` with curated reset** (`RESET ALL; RESET
  SESSION AUTHORIZATION; UNLISTEN *; CLOSE ALL;`). DISCARD's
  `DEALLOCATE ALL` killed tokio_postgres' typeinfo cache, producing
  intermittent `prepared statement "sN" does not exist` errors on
  custom-type queries after cached-conn reuse. New regression tests:
  `test_postgresql_custom_types_on_cached_connection` and
  `test_postgresql_set_role_does_not_leak_across_cached_connection`
  (the latter catches the case where someone might switch back to
  `RESET ALL` alone and silently introduce a permission-leak vector —
  RESET ALL doesn't cover SET ROLE / SET SESSION AUTHORIZATION).

- **ISO-8601 timestamp results** (`pg_cell_to_json_value`). Pre-fix
  `TIMESTAMP` was rendered with a space separator ("2024-01-15 10:30:00")
  and `TIMESTAMPTZ` with " UTC" suffix ("2024-01-15 10:30:00 UTC") —
  neither parseable by `date-fns parseISO`, JavaScript `new Date()` is
  lenient enough to handle them but several frontend `App*Input.svelte`
  components use parseISO and fail silently. Switched to ISO-8601 with
  `T` separator and `+00:00` offset; arg-parsing path still accepts the
  legacy " UTC" suffix for back-compat.

Test coverage:
- 17/17 unit (`pg_executor::tests`)
- 9/9 integration (`backend/tests/worker.rs`, `test_postgresql_*`)
- 27/27 parser (`windmill-parser-sql`)
- 42/42 SDK (`typescript-client/tests/sqlUtils.test.ts`)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(pg): bounded one-shot warning on numeric precision loss + ISO-8601 + NaN handling

Found while probing PG-script DX with millions of numeric cells:

1. **Numeric precision-loss warning**: `numeric` results are still serialised
   as JSON Number (back-compat — switching to JSON String would silently
   break user code doing arithmetic on results), but we now detect
   `Decimal -> f64 -> Decimal` round-trip failure and emit a single
   job-log warning recommending a `::text` cast in the SQL. Bounded by
   `NUMERIC_PRECISION_CHECK_BUDGET = 256` cells per query (one atomic
   load + one fetch_sub on the hot path; first lossy value
   short-circuits to a single load thereafter). Worst-case overhead on
   a 1M-cell numeric-heavy query: ~25µs of checks + 5ns × N atomic
   loads (vs. ~100ms unbounded).

2. **ISO-8601 timestamps**: `pg_cell_to_json_value` previously returned
   `"2024-01-15 10:30:00"` (TIMESTAMP) and `"2024-01-15 10:30:00 UTC"`
   (TIMESTAMPTZ) — neither parseable by date-fns `parseISO`, which is
   what the apps `App*Input.svelte` components use, so timestamp values
   silently failed to round-trip into date pickers. Switch to ISO-8601
   (`T` separator + `+00:00` offset) on the result side; arg-parser
   continues to accept the legacy `" UTC"`-suffixed format for
   back-compat.

3. **Float NaN / Infinity results**: `Number::from_f64` returns None for
   NaN / ±Inf, which `pg_cell_to_json_value` was raising as
   "invalid json-float" — failing the *entire* query if any cell held
   one of these special values. Now serialise them as JSON strings
   ("NaN", "Infinity", "-Infinity") and let the rest of the row come
   through. Arg-side: `s.parse::<f64>()` already accepts the same
   strings.

Tests:
- `decimal_fits_f64_losslessly_predicate` — covers fits / doesn't-fit
  cases for the precision-loss predicate.
- `precision_check_budget_caps_per_query_overhead` — locks in the
  budget cap and the loss-flag short-circuit.
- All 9 PG integration tests + 17 unit tests pass.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(pg): add pg_advisory_unlock_all to reset; warn on missing args; honor decl defaults

While probing PG-script DX further found three more frictions:

1. **Advisory lock leak** (cubic P2): switching from `DISCARD ALL` to
   `RESET ALL; RESET SESSION AUTHORIZATION; UNLISTEN *; CLOSE ALL;`
   meant session-scoped advisory locks (`pg_advisory_lock`) leaked
   across cached-connection reuse. Add `SELECT pg_advisory_unlock_all()`
   to the chain — `DISCARD ALL` covered this implicitly via
   `DISCARD PLANS / DEALLOCATE / pg_advisory_unlock_all` and we lost it
   in the switch.

2. **Missing-arg silent NULL**: an arg declared in the SQL (e.g.
   `-- $1 amount (numeric)`) but not provided in the args object was
   bound as NULL with no error / warning. Misspelling the key in the
   args object silently produced a row of NULLs — a notorious DX
   debugging trap. Now: collect the names of declared-but-missing
   args during dispatch and emit a single one-shot warning to the job
   logs at end-of-query naming each one. Bound NULL is preserved for
   back-compat.

3. **Declaration defaults ignored**: `-- $1 a (int) = 5` carries
   `arg.default = Some(Number(5))`, but the dispatch fell straight to
   NULL when the arg was missing. Now: respect the default —
   user-supplied value > declaration default > NULL. Also fixes the
   warning logic above (only warn for args that *don't* have a default).

Tests: existing 19 unit + 9 integration pass.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(pg): multi-word PG types with [] suffix lost the array-ness; array arms accept stringified values

Two more frictions found while probing SDK end-to-end against a real
datatable resource:

1. **Multi-word array types lose the [] suffix in the parser**.
   `transform_types_with_spaces` recognises aliases for "double
   precision", "character varying", "timestamp with time zone", etc.
   but its return type was `&'a str` — only the bare alias, never with
   a trailing `[]`. The `RE_CODE_PGSQL` regex's `\w+` captures stop at
   the first space, so the regex's own `(?:\[\])?` array-suffix branch
   sees only `"double"` (not `"double precision[]"`); the `[]` was
   silently lost. Result: `$1::double precision[]` (which the SDK now
   emits for homogeneous float arrays via the new auto-tag) routed
   through `Value::Array → Type::JSONB` and the server failed with
   "cannot cast type jsonb to double precision[]".

   Fix: switch `transform_types_with_spaces` to return `Cow<'a, str>`
   and re-check the trailing bytes after a multi-word match. If they
   start with `[]`, return `format!("{alias}[]")` — Owned. Single-word
   types and the no-match path keep returning Borrowed slices, so no
   allocation in the hot path.

2. **Array arms in `convert_vec_val` rejected stringified values for
   numeric / int* / bool / oid / real / double**. The scalar `convert_val`
   already parses strings into the matching native type for these arg_ts,
   but the array variant only accepted JSON-native counterparts. Sending
   `["1.5", "2.5", "3.5"]` against `$1::numeric[]` (e.g. via `unnest` for
   bulk loading, or `JSON.stringify(BigInt[])` round-trip) failed with
   "Mixed types in array". Now the array arms mirror the scalar ones —
   `as_<native>().or_else(|| as_str().and_then(parse))` — so both shapes
   round-trip cleanly.

Tests: 19 unit + 9 integration pass; existing parser tests cover the
multi-word array forms (the regex-cap behaviour didn't break for
single-word types, and Cow plumbing is transparent to all callers).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(parsers): add otyp_inferred field to Arg literals in tests + 3 missed src files

CI failures: the perl-driven sweep that added `otyp_inferred: false` to
every `Arg { ... }` literal when I introduced the field in the parser
schema covered `src/lib.rs` files but missed:

  - parsers/windmill-parser-bash/src/lib.rs       (mass-edited but a
    later format pass un-applied a few sites)
  - parsers/windmill-parser-go/src/lib.rs         (same)
  - parsers/windmill-parser-graphql/src/lib.rs    (same)
  - parsers/windmill-parser-nu/tests/tests.rs     (test file — not
    swept the first time)
  - parsers/windmill-parser-ts/tests/tests.rs     (test file — same)

Also tightened the regex to handle `oidx: None` without the trailing
comma (some test files had the field as the last initialiser line).

`cargo build --features <CI feature combo> --workspace --all-targets`
is clean.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(sdk): Date → TIMESTAMPTZ; NaN / ±Infinity → string

Two more frictions found while running the actual SDK end-to-end against
a live datatable resource:

1. **JS `Date`** fell into the typeof "object" branch and was tagged
   `::JSON`. It worked accidentally for `${date}::timestamptz` via PG's
   `json → text → timestamptz` implicit cast chain, but `${date}` against
   a `timestamptz` column without a user-supplied cast bound the value
   as a JSON string and the comparison `timestamptz = json` failed. Now:
   `inferSqlType` recognises `Date` and tags `::TIMESTAMPTZ`;
   `serializeArgValue` emits `Date.toISOString()` so the executor's
   `Value::String → TIMESTAMPTZ` arm parses it cleanly.

2. **JS `NaN` / `±Infinity`** silently became NULL. `JSON.stringify(NaN)`
   returns `"null"` per the JS spec, so the value reached the executor as
   JSON null — the SDK's `::DOUBLE PRECISION` tag then bound a NULL
   double. Fix: detect non-finite numbers in `serializeArgValue` and
   stringify them as `"NaN" / "Infinity" / "-Infinity"`. The executor's
   `Value::String → FLOAT8` arm (`f64::from_str`) accepts these literals
   directly, and the result-side already renders the values as JSON
   strings (matching round-trip).

SDK unit tests grow from 42 → 44 passing.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* test(pg): integration coverage for multi-word arrays + stringified array elements

Locks in the two array fixes from the previous commit
(`fix(pg): multi-word PG types with [] suffix lost the array-ness`)
with end-to-end cases in `test_postgresql_arg_type_combinations`:

- `double precision[]`, `character varying[]`, `timestamp without time
  zone[]` — verifies the parser keeps the `[]` suffix after multi-word
  alias resolution.
- `numeric[]` / `int[]` / `bool[]` from stringified primitives — verifies
  the array arms of `convert_vec_val` apply the same string-coercion
  the scalar arms do.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* style: fix indentation drift on otyp_inferred lines

cargo fmt cleanup of leftover indentation where the perl-driven sweep
that introduced the otyp_inferred field landed at the wrong column.
No behaviour change.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
 2026-05-01 17:08:59 +00:00
centdixandClaude Opus 4.5 9a60ff2e77 feat: add ai agent conversation output control (#8915) 
* feat: add ai agent chat output flag

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: suppress ai agent tool chat messages

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* refactor: rename ai agent conversation output flag

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* feat: expose ai agent conversation output toggle

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: gate ai agent chat tab by chat mode

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* chore: regenerate system prompts

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: address ai agent chat review feedback

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
 2026-04-23 14:13:20 +00:00
centdixandClaude Opus 4.6 26a6d1e4ce refactor: create windmill-ai crate (part 1 — types, traits, base modules) (#8530) 
* refactor: create windmill-ai crate and move base AI types from windmill-common

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: move worker AI types to windmill-ai crate

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: move QueryBuilder trait and StreamEventSink abstraction to windmill-ai

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix: add base64 dependency to windmill-ai for bedrock PDF support

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* docs: add windmill-ai refactor plan

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: address PR review — remove dead bedrock feature, add boxed_sink helper, move plan to docs

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-22 11:20:27 +00:00
centdixandClaude Opus 4.5 b1778272fc fix: clean ai memory and cache bedrock prompts (#8847) 
* fix: avoid persisting system prompts in ai memory

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: keep ai memory cleanup write-side only

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* feat: add bedrock prompt caching for claude

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* test: add bedrock memory regression

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: gate bedrock prompt caching by model id

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* docs: link bedrock caching allowlist source

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
 2026-04-16 14:30:22 +00:00
Ruben FiszelClaude Opus 4.6windmill-internal-app[bot]
 		5b3913052e refactor: convert read-hot globals to AtomicBool/I64 and ArcSwap (#8815) 
* refactor: extract load helpers from reload_setting family

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: convert atomic primitive globals to AtomicBool/AtomicI64

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: convert CRITICAL_*/HUB_API_SECRET/INSTANCE_EVENTS_WEBHOOK/JWT_SECRET to ArcSwap

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: pin ee-repo-ref to arcswap-refactor EE branch commit

* refactor: convert BASE_URL/HUB_BASE_URL/MIN_VERSION/LICENSE_KEY*/LICENSE_KEY_ID to ArcSwap

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: convert worker hot-path globals to ArcSwap (WORKER_CONFIG et al)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: pin ee-repo-ref to combined arcswap-urls+worker EE commit

* chore: update ee-repo-ref to d8be8f88cb8898c8f6b27421989d53528223815d

This commit updates the EE repository reference after PR #532 was merged in windmill-ee-private.

Previous ee-repo-ref: c375aaaac9ec0fc0480993627d0defc8054c31a4

New ee-repo-ref: d8be8f88cb8898c8f6b27421989d53528223815d

Automated by sync-ee-ref workflow.

* fix: cleanup unused imports + fix 2 missed WORKER_CONFIG readers

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: update ee-repo-ref to ce0f8fbbbde09c4a858312d2d8716d224e99042c

This commit updates the EE repository reference after PR #534 was merged in windmill-ee-private.

Previous ee-repo-ref: 450b601b5aba0ca0b2045f4b5071aa8701b4bfb7

New ee-repo-ref: ce0f8fbbbde09c4a858312d2d8716d224e99042c

Automated by sync-ee-ref workflow.

* fix: secret_backend_integration test — BASE_URL.write().await → .store()

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: convert APP_WORKSPACED_ROUTE to AtomicBool for symmetry with HTTP_ROUTE_WORKSPACED_ROUTE

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: update ee-repo-ref to e587df8 (post-#535 merge)

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
 2026-04-14 00:04:10 +00:00
Ruben FiszelClaude Opus 4.6windmill-internal-app[bot]
 		2d18a68099 feat: add scheduled job deletion with configurable retention period (#8753) 
* feat: add scheduled job deletion with configurable retention period

Extends delete_after_use with delete_after_secs to enable configurable
retention periods for job args/result/logs. At completion, jobs can be
scheduled for future deletion via a new job_delete_schedule table,
processed by a monitor task. Supports per-script, per-flow, and
per-flow-step configuration. Backward compatible.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* feat: add integration tests, revert query! macros, fix review issues

- Add integration tests for resolve_delete_after_secs, schedule_job_deletion,
  flow-level and module-level delete_after_secs, backward compat
- Revert sqlx::query() back to sqlx::query!() macros for compile-time safety
- Regenerate sqlx offline cache
- Fix FlowModule/NewScript/FlowValue constructions in all test files
- Fix autoscaling_ee.rs for updated script_path_to_payload return type

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: update ee-repo-ref.txt for autoscaling_ee fix

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix: gate cleanup_scheduled_job_deletions behind enterprise feature

Prevents dead_code warning (which CI treats as error via -D warnings)
when compiling without enterprise feature.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: regenerate sqlx cache after merge with main

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix: address review feedback on scheduled deletion

- Monitor: roll back transaction on any cleanup error so schedule rows
  survive for retry on next cycle (instead of best-effort then discard)
- Migration: add FK with ON DELETE CASCADE to job_delete_schedule.job_id
  to prevent orphan rows when jobs are deleted through other means
- Simplify bool-to-Option conversion with .then_some(true)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: stop setting delete_after_use alongside delete_after_secs

No mixed-version deployment scenario exists, so delete_after_secs alone
is sufficient. The backend's resolve_delete_after_secs handles
(None, Some(secs)) correctly without needing delete_after_use set.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: remove delete_after_use from public API surface

Remove delete_after_use from OpenAPI spec, API client, runtime client,
and workspace export. Only delete_after_secs is exposed going forward.

The field remains in Rust backend types with #[serde(skip_serializing)]
for backward-compatible deserialization of existing scripts/flows that
were saved with delete_after_use: true.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: update ee-repo-ref to 1d4b7a31fc115d6aba8640f7cd3fd5a01abe6806

This commit updates the EE repository reference after PR #519 was merged in windmill-ee-private.

Previous ee-repo-ref: 9eba09a13b778caafc6ae65098b90e53c91984d3

New ee-repo-ref: 1d4b7a31fc115d6aba8640f7cd3fd5a01abe6806

Automated by sync-ee-ref workflow.

* fix: regenerate system prompts, remove unused import

- Regenerate auto-generated system prompts after openflow schema change
- Remove unused serde_json::json import in test file (CI -D warnings)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix: insert dummy v2_job row in schedule tests for FK constraint

The job_delete_schedule table has a FK to v2_job, so tests need a
real v2_job row before inserting into the schedule table.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: trigger CI re-run

* fix: remove heavy flow integration tests to avoid CI worker contention

The flow integration tests spawn workers that compete for CPU with
the existing relock_skip tests under --test-threads=10, causing
consistent 60s timeouts in CI. Keep only the lightweight unit tests
and DB integration tests.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix: restore correct ee-repo-ref for our branch

The ref was overwritten to main's EE ref during a rebase. Restore to
our branch's EE commit that includes the autoscaling tuple fix.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* chore: retrigger CI on fresh runner

* fix: remove FK constraint from job_delete_schedule to unblock CI

The FK with ON DELETE CASCADE to v2_job may have caused performance
overhead during test DB setup (each sqlx::test creates a fresh DB
with all migrations). Remove the FK — orphan schedule rows are
harmlessly cleaned by the monitor.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* ee-ref

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
 2026-04-08 04:15:28 +00:00
centdixandClaude Opus 4.6 e44504c6e9 feat: add PDF input support to AI agent (#8525) 
* feat: add PDF input support to AI agent with user_attachments field

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* test: add integration tests for PDF input and backward compat

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* feat: add ContentPart::File variant for PDF support across all providers

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* refactor: address review feedback on PDF support

- Extract parse_data_url_bytes and mime_to_document_format helpers in Bedrock
- Add is_document_mime helper in ai_types for centralized MIME routing
- Extract s3_object_to_content_part helper to deduplicate image_handler/openai
- Rename AnthropicImageSource to AnthropicBase64Source
- Derive Bedrock DocumentFormat from MIME type instead of hardcoding Pdf

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* fix: merge user message and attachments into single message for Bedrock

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-26 13:55:10 +00:00
hugocasaandClaude Opus 4.6 f4489cbe64 fix: prevent AI agent tool jobs from becoming zombies on cancellation (#8437) 
* fix: prevent AI agent tool jobs from becoming zombies on cancellation

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* kill tool tasks on cancel timeout

* fix: address review feedback and update sqlx cache

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-18 18:30:31 +00:00
centdixandClaude Opus 4.6 5acb367cf9 feat: support custom headers in customai resource type (#8364) 
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-16 11:31:22 +00:00
centdixClaude Opus 4.6hugocasa
 		cb349cb3d1 feat: add Vertex AI support for Google Gemini models (#8303) 
* refactor: rename AnthropicPlatform to AIPlatform for generic vertex support

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: add Vertex AI support for Google Gemini models

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* chore: update platform doc comments to be provider-generic

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* chore: update googleai default models to latest gemini 2.5 and 3

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: move AIPlatform to windmill_common to avoid duplication

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: hugocasa <hugo@casademont.ch>
 2026-03-10 18:20:19 +00:00
Ruben FiszelandClaude Opus 4.6 a6d4390790 feat: workflow-as-code (WAC) v2 (#8172) 
* feat: workflow-as-code v2 with @task decorator API

Replace ctx.step("name", "script") API with @task decorators where
functions are called directly. Users no longer need to pass WorkflowCtx
or use string-based step names/script paths.

Python: @task decorator with contextvars-based implicit context
TypeScript: task() wrapper with module-level context variable
Parsers: detect @task function calls instead of ctx.step() calls
Worker: updated wrappers to set implicit context

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: WAC v2 checkpoint/replay with _executing_key child dispatch

- Rust-side orchestration: parent dispatches child jobs, suspends, resumes on completion
- _executing_key in checkpoint tells child which step to execute directly
- task() throws StepSuspend(mode="step_complete") after executing target step
- result_processor handles child completion and updates parent checkpoint
- WacGraph.svelte for runtime execution visualization
- Sequential and parallel workflows tested end-to-end

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: WAC v2 bundle cache, globalThis ctx sharing, description optional

- Disable bun bundle caching for WAC v2 scripts (wrapper needs
  windmill-client from node_modules, not available in bundle mode)
- Use Reflect.set/get(globalThis, "__wmill_wf_ctx") to share workflow
  context across dual module instances (wrapper vs user script)
- Never-resolving thenable for non-matching steps in child job mode
  prevents Promise.all race conditions
- Make description field optional in NewScript API (defaults to "")

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: add step() primitive for inline checkpointed steps

step() executes a function inline (no child job) and persists the result
to the checkpoint. On replay, the cached value is returned — ensuring
deterministic behavior for non-deterministic operations like Date.now()
or Math.random().

- TypeScript: step(name, fn) — executes inline, throws StepSuspend with
  mode "inline_checkpoint" to persist before continuing
- Rust: InlineCheckpoint variant in WacOutput, saves to checkpoint and
  resets running=false for immediate re-pickup (no zombie wait)
- Shared step counter between task() and step() via _allocKey()

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: add Python WAC v2 support with task(), step(), workflow()

- Python SDK: WorkflowCtx with _executing_key child mode, _alloc_key
  shared counter, _run_inline_step for step(), _execute_directly and
  _never_resolve for child mode, step() async function
- Python executor: WAC v2 detection, checkpoint.json writing, WAC
  wrapper.py generation calling _run_workflow(), post-execution hook
  into shared handle_wac_v2_output()
- Make handle_wac_v2_output pub so both bun and python executors share
  the same dispatch/suspend/inline-checkpoint logic
- 17 Python tests covering dispatch, replay, parallel, conditional,
  inline checkpoint, and child mode

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* chore: update sqlx prepared queries

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: WacGraph Tooltip→Popover, simplify wacToFlow parsers

- Fix type error: Tooltip doesn't accept text snippet, use Popover
- Extract shared helpers for task matching and block collection
- Replace linear tasks.find() with Map lookups
- Remove mutable module-level counter

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: Box::pin WAC v2 output handler to prevent stack overflow

handle_python_job's async state machine was too large when combined
with handle_wac_v2_output. Box::pin heap-allocates the future.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: merge WAC v1 and v2 task decorators to preserve backward compat

The v2 @task decorator was shadowing the v1 one, breaking WAC v1
scripts that rely on HTTP-based dispatch via /workflow_as_code/ API.

The merged decorator handles three modes:
- v2: inside @workflow context → checkpoint/replay dispatch
- v1: WM_JOB_ID set, no @workflow → HTTP API dispatch + wait_job
- standalone: no Windmill env → execute function body directly

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: skip no_main_func detection for WAC v2 scripts in TS and Python parsers

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: prevent empty/noop dispatch causing infinite requeue loop

- Validate steps.len() > 0 in WAC dispatch handler (issue 3)
- Replace noop StepSuspend throw with never-resolving promise so it
  can't reach the backend as an empty dispatch (issue 4)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: Python task wrapper now converts positional args to kwargs in v2 mode

Previously only **kwargs were passed to _next_step(), silently dropping
positional arguments. Extract shared _merge_args() helper used by both
v1 and v2 paths.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: replace unwrap() with proper error propagation in WAC arg serialization

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: add workspace_id filter to v2_job queries in WAC dispatch

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: prevent race condition in WAC child dispatch

Restructure dispatch to save checkpoint + suspend parent + seed child
checkpoints in a single transaction BEFORE pushing child jobs. This
ensures a fast child can't complete before the parent is suspended.

Also wrap InlineCheckpoint save + running reset in a transaction to
prevent corrupted state on crash.

Use ULID for pre-generated child job IDs (consistent with rest of API).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: include step key and child job ID in WAC error propagation

Move step_key lookup before the success check so failed child errors
include which task failed, the child job ID, and the original error.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* docs: document WAC determinism contract and step dispatch semantics

- Document that workflow functions must be deterministic across replays
- Document that WacStepDispatch.script/args are metadata, not dispatch targets
- Add comments on counter-based key allocation

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: tighten WAC v2 detection to reduce false positives

Replace naive substring matching with line-aware checks that skip
comments and look for specific patterns:
- TS: import from "windmill-client" containing workflow/task
- Python: @workflow and @task decorators with wmill import

Extracted shared helpers in wac_executor.rs used by both executors.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: show failed steps in WacGraph when workflow completes with errors

When flowDone is true and a pending step isn't in completedSteps,
mark it as 'failed' instead of 'running'. The failed state CSS and
XCircle icon were already defined but never triggered.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: unsuspend and fail parent when WAC child push fails

Previously if a child push failed mid-batch, the parent remained
suspended with suspend = num_steps but fewer children, hanging until
the 14-day timeout. Now the push loop catches errors and unsuspends
the parent before returning the error.

Also adds source hash validation: if the script content changes between
replays, the job fails with a clear error instead of silently feeding
stale checkpoint data into wrong steps.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: clear suspend_until when unsuspending WAC parent

Set suspend_until = NULL alongside suspend = 0 in both the child
failure and all-children-complete paths, so the parent doesn't rely
on subtle pull query invariants to be re-picked-up.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* test: add exhaustive edge case tests for WAC v2 SDK

fix: make TS task wrapper non-async to fix unawaited task flush

The async wrapper caused microtask-based thenable auto-resolution that
fired .then() and threw StepSuspend before _flushPending() could capture
unawaited steps — making the flush mechanism completely broken. Now the
thenable is returned directly without async wrapping. Backward compatible
with v1 (all code paths still return awaitables).

Tests added (59 TS + 66 Python) covering: full sequential lifecycle,
step after parallel, parallel after parallel, conditional on step result,
empty/single-task workflows, 10+ steps, falsy value preservation, inline
steps, mixed step/task, unawaited flush, child mode with parallel,
key determinism, large parallel groups, and complex mixed patterns.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: atomic checkpoint updates to prevent parallel child race condition

Replace read-modify-write pattern in handle_wac_child_completion with
atomic SQL operations:
- completed_steps merged via jsonb_set(... || jsonb_build_object(...))
  so concurrent children on different workers don't overwrite each other
- suspend counter decremented atomically with RETURNING to determine
  "all done" condition (instead of checking completed_steps in memory)
- suspend_until cleared in the same atomic decrement statement

Before this fix, two parallel children completing simultaneously could
both load the same checkpoint, each add their step, and save — the
second write would overwrite the first, silently losing a child result
and leaving the parent suspended forever.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: cancel already-pushed children on partial WAC dispatch failure

When pushing child jobs sequentially, if pushing child N fails, children
1..N-1 are already running. Previously the error handler only unsuspended
the parent, leaving orphaned children that would complete and corrupt the
checkpoint state (decrementing suspend on an already-unsuspended parent,
potentially causing duplicate step execution on re-run).

Now on partial failure:
1. Cancel all already-pushed children (prevents them from completing
   and corrupting checkpoint state)
2. Clear pending_steps from checkpoint (so parent doesn't think
   children are outstanding on re-run)
3. Then unsuspend parent (so the error propagates)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: skip WAC duration write and child check for non-WAC parents

The duration write to workflow_as_code_status was running for every
non-flow child with a parent (error handlers, success handlers,
run_script children), even though it was only intended for WAC jobs.

Add WHERE workflow_as_code_status IS NOT NULL to skip non-WAC parents
entirely. Piggyback RETURNING pending_steps.job_ids on the same query
so WAC v2 child completion needs zero extra DB round-trips on the
success path.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: seed child checkpoint in same transaction as push

The child checkpoint insert was happening before the child job was
pushed, violating the FK constraint on v2_job_status. Move it into
the push transaction so the job row exists and the child can't be
picked up before its checkpoint is ready.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: set running=false when WAC parent suspends for child dispatch

The parent job kept running=true after suspending, so workers wouldn't
pick it up when children completed and suspend reached 0. The parent
only advanced when the zombie job detector reset it (~90s). Now the
dispatch suspend sets running=false so the parent is immediately
eligible for pickup.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: WAC parent suspend/unsuspend lifecycle

Keep running=true when suspending the parent so the normal pull query
(WHERE running=false) never picks it up. Keep suspend_until non-null
when decrementing suspend to 0 so the suspended pull query
(WHERE suspend_until IS NOT NULL AND suspend<=0) picks it up.

Previously: setting running=false caused infinite restart loops because
the normal pull query has no suspend check and would immediately re-pick
the parent. Clearing suspend_until on the last child prevented the
suspended pull from ever seeing it, requiring the 90s zombie detector.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: add approval primitive, flow child completion, timeline fixes for WAC v2

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: add error propagation, task options, sleep, and parallel for WAC v2

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* test: fix python SDK tests to use name-based keys and add new test coverage

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: address WAC v2 review findings (sleep timing, error marker, atomicity)

- Fix sleep using suspend=1 instead of 0 to enforce actual delay
- Add approval/sleep resume injection to Python executor
- Fix TS SDK concurrency_limit mapping (was reading wrong property)
- Namespace error marker as __wmill_error to avoid user data collision
- Wrap child completion SQL in transaction for atomicity
- Decrement suspend even when step key is missing (prevents hang)
- Expand TASK_RE to handle export const, let, var, generics
- Validate step key uniqueness before dispatch
- Log warning on checkpoint deserialization failure
- Remove unimplemented delete_after_use from SDKs
- Add TaskError exception class to Python SDK with diagnostic context
- Fix extra positional args handling and add functools.wraps
- Improve getParamNames to handle typed/destructured params

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* sqlx

* sqlx

* test: add WAC v1 e2e integration tests for TS and Python

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: revert fake test versions in typescript-client

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: remove unused WacGraph component and strip wacToFlow to isWorkflowAsCode

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: extract shared approval/sleep resume logic into wac_executor

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-09 19:39:24 +00:00
centdixClaude Opus 4.6hugocasa
 		825df2161e refactor: extract google ai logic to windmill-common and use native gemini api in chat proxy (#8115) 
* refactor: extract google ai logic to windmill-common and use native gemini api in chat proxy

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix: use x-goog-api-key header for google ai non-chat requests

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix: transform gemini models response to openai format and use correct auth header

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix: skip thought parts from gemini thinking models in sse stream

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* Revert "fix: skip thought parts from gemini thinking models in sse stream"

This reverts commit dfa01d282c.

* fix: handle tool calls and sanitize schemas in gemini chat proxy

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* refactor: move Gemini→OpenAI response conversion to windmill-common

Extract streaming and non-streaming Gemini response conversion into
shared functions in ai_google so the API proxy and worker use the same
logic instead of duplicating format translation.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: review fixes for google ai refactor

- Remove duplicate parse_data_url from worker utils, use shared version
  from windmill_common::ai_google in both google_ai and anthropic providers
- Improve error diagnostics in google.rs by including HTTP status code
  in error messages from Gemini API responses
- Change GeminiToolCallEvent::into_extra_content to instance method
  to_extra_content using &self

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: deduplicate worker Gemini message conversion using pre-flight pattern

Replace the worker's `convert_messages_to_gemini` and
`convert_content_to_parts_with_s3` (~130 lines) with the existing
pre-flight pattern: `prepare_messages_for_api` converts S3 objects to
data URLs, then the shared `openai_messages_to_gemini` handles the rest.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: hugocasa <hugo@casademont.ch>
 2026-03-09 15:15:37 +00:00
centdixandClaude Opus 4.6 de6fd160d5 feat(aiagent): handle ai agent as tool (#8031) 
* worker: support AIAgent tools in AI executor

* worker: complete nested AIAgent tool execution path

* worker: inline AIAgent tool schema usage

* fix agent action

* frontend: add AI Agent as tool type in flow builder

Add the ability to insert a nested AI Agent as a tool within another
AI Agent step. Includes type definitions, factory function, graph icon,
insert/event wiring, and a dedicated editor component.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: remove AiAgentToolEditor, reuse FlowModuleComponent for AI agent tools

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: populate all input transforms for nested AI agent tools

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: avoid missing v2_job_status error for nested AI agent tools

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* sqlx

* nit

* refactor: cleanup nested AI agent tool implementation

- Add max nesting depth guard (5) on parent chain traversal
- Reject 3+ level nesting explicitly with clear error message
- Remove unnecessary flow_step_id tuple scaffolding in tool dispatch
- Consolidate get_value() calls using borrow in first match
- Replace unsafe `as unknown as FlowModule` casts with agentToolToFlowModule()
- Simplify toolKind ternary chain with .includes() lookup
- Fix leftover over-indentation from tuple removal
- Remove duplicate doc comment on is_completed_input_transform

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: pass flow_step_id and flow_job_id overrides to run_agent for nested AI agents

For nested AI agent tools, job.flow_step_id is None and job.parent_job
points to the parent agent instead of the flow. This caused memory
read/write and flow context resolution to silently fail.

handle_ai_agent_job already computes the correct flow_step_id (via
runnable_path fallback) and flow_job_id (via parent chain traversal).
This change threads those values through run_agent and
ToolExecutionContext so all downstream consumers use the correct IDs.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* cleaning

* cleaning

* move const

* fix

* refactor: replace defaultToAi boolean with allowedAiTransforms whitelist

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* refactor: propagate root_job at push time, remove flow_job_id_override

Instead of threading flow_job_id_override through run_agent and
get_flow_context, propagate root_job and flow_innermost_root_job
when pushing tool jobs so nested AI agents can find the flow
job naturally via the existing job fields.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: simplify nested AI agent parent chain walk-up

Replace the generic depth-limited loop with a single-level check since
only flow → agent → nested agent tool is supported. Remove
MAX_AGENT_NESTING_DEPTH constant and flatten the module lookup.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: reject 3+ level nested AI agent tools before job creation

Check at the parent agent level whether a nested AIAgent tool contains
AIAgent sub-tools. If so, return a fatal error immediately, preventing
the sub-job from being created and avoiding retry loops.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: resolve deadlock in nested AI agent tool execution

Replace channel forwarding with inline DB writes for tool job
completions. Nested agents used bounded(1) channels where a sub-tool's
forwarded result would fill the parent channel, leaving no room for the
agent's own completion — causing a deadlock. Writing directly via
add_completed_job/add_completed_job_error bypasses the channel entirely.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-23 15:59:39 +00:00
Ruben FiszelandClaude Opus 4.6 6bf544f507 refactor: extract object store into dedicated crate with filesystem backend (#7996) 
* refactor: extract object store code into windmill-object-store crate with filesystem backend

Consolidate all object_store-dependent code from windmill-common into a new
windmill-object-store crate. Add a filesystem-backed object store implementation
using LocalFileSystem for dev/testing without cloud credentials. Includes 30
comprehensive tests covering render_endpoint, lfs_to_object_store_resource,
duckdb_connection_settings, error mapping, and filesystem-backed integration tests.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* all

* all

* all

* all

* fix: fix raw_app hardcoded path, add missing ObjectStoreResource import, and add tests

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: move S3ModeFormat to windmill-types, make windmill-parser-sql optional, restore debug logs

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* all

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-18 12:24:45 +00:00
centdix d95e4db8f3 feat(ai): add AWS bedrock session token support (#7908) 
* Add AWS Bedrock session token support in API and worker

* Add Bedrock auth mode integration tests for AI agents

* Split Bedrock integration test env vars for IAM and session creds

* cleaning

* Add masked Bedrock bearer-token debug info logs

* Revert "Add masked Bedrock bearer-token debug info logs"

This reverts commit 6b2fc5e7c2d5b1c6db81f416a4439941a084108c.

* cleaning
 2026-02-11 19:29:58 +00:00
Devdatta Talele 5effb87a36 fix: strip unsupported schema fields for Google AI (#7894) 
Closes #7759
 2026-02-11 13:26:08 +00:00
centdixandClaude Opus 4.6 f22eb964e4 feat(ai): support 1M context window for Anthropic resources (#7891) 
* feat(ai): support 1M context window for Anthropic resources

Add `enable_1m_context` boolean field to Anthropic resource configuration.
When enabled (and not using Vertex AI), sends the `anthropic-beta: context-1m-2025-08-07`
header in both the API proxy layer and the AI agent worker layer.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix(ai): add serde alias for enable_1M_context DB field name

The resource_type schema uses `enable_1M_context` (uppercase M) but
serde only matched `enable_1m_context` and `enable1mContext`, causing
the field to always deserialize as false.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-11 12:01:46 +00:00
centdixandClaude Opus 4.5 6272cd17a4 feat(aiagent): add prompt caching for Anthropic models (#7878) 
* feat: add prompt caching support for Anthropic API

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* exclude vertex

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
 2026-02-10 17:42:49 +00:00
Ruben FiszelandClaude Opus 4.6 ff70a4e9d1 fix: parse Python datetime.datetime and datetime.date type annotations (#7856) 
* fix: parse Python datetime.datetime and datetime.date type annotations correctly

The Python parser only matched ExprKind::Name for type annotations, so
`datetime.datetime` (an Attribute expression) silently fell through to
Typ::Unknown and no datetime picker was shown in the UI.

- Extend parse_expr to resolve `datetime.*` attribute access (alongside
  the existing `wmill.*` handling)
- Add Typ::Date variant for `datetime.date` → JSON schema format "date"
- Update python worker to import and convert `date.fromisoformat()`
- Update argSigToJsonSchemaType, AI types, schema validation, and SQL
  datatype wasm for the new Date variant

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* all

* all

* all

* all

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-09 15:17:26 +00:00
Ruben FiszelandClaude Opus 4.6 e1a815f6a0 refactor: extract windmill-dep-map crate for parallel api/worker compilation (#7846) 
* refactor: extract windmill-dep-map crate for parallel api/worker compilation

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: resolve WebhookShared type mismatch and missing enterprise propagation

- Make windmill-api webhook_util re-export from windmill-common instead of
  duplicating types, fixing Extension<WebhookShared> mismatch between
  windmill-store and windmill-api
- Add windmill-api-jobs/enterprise to windmill-trigger enterprise feature
  so check_license_key_valid is available when trigger subcrates enable
  enterprise on windmill-trigger

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: stop trigger features from unconditionally enabling enterprise

Move enterprise propagation for all trigger subcrates from individual
trigger feature definitions to the enterprise feature itself, so
enterprise is only enabled when explicitly requested.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* refactor: remove unused pub use re-exports and disable CI cargo cache

- Remove unused re-exports from windmill-worker/src/lib.rs:
  trigger_dependents_to_recompute_dependencies, handle_job_error,
  and unused bun/otel items
- Fix callers to use direct module paths instead
- Add windmill-dep-map as dev-dependency for tests
- Disable cargo cache in backend-check CI (faster from-scratch builds)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: restore bun re-exports used by tests

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* all

* chore: re-enable cargo cache for check_ee_full CI job

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-08 00:39:56 +00:00
centdixandClaude Opus 4.5 18d85f1412 fix: remove $schema field from Google AI output schema requests (#7765) 
* fix: remove $schema field from Google AI output schema requests

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* test: add $schema field to all output schema integration tests

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* fix: remove $schema field from Google AI tool parameter schemas

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* test: add workspace script tool test for AI agents

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
 2026-02-03 01:04:36 +00:00
Diego ImbertDiego Imbertclaude[bot]claude[bot]windmill-internal-app[bot]
 		635a24f82c feat: runtime assets (#7656) 
* Runtime assets

* Nits

* Revert "Nits"

This reverts commit 3031a2ddd1.

* detection_kinds

* don't delete runtime assets

* Show latest executions

* conditional unique idx

* nit status

* refactor

* nit refactor

* prepare sql

* Detect assets in complex JSON input objects

* false positive prevent

* nit

* redundant idx

* Update frontend/src/lib/components/assets/AssetsUsageDrawer.svelte

Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>

* Update backend/migrations/20260122134517_runtime_assets.up.sql

Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>

* runtime assets are inserted in a loop

* nit

* nit fix

* Don't use lazy static

* fix compilation

* nits

* missing on conflict do nothing

* add index

* Fix max n logic

* created at

* nits

* remove pagination

* sqlx prepare

* Only detect resource assets in input

* get_runtime_asset_sender()

* use global get_runtime_asset_sender to avoid prop drilling

* nit refactor : register_runtime_asset

* get job_id from token

* job as a usage kind

* fixes

* ee

* nit refactor

* merge access types when same job uses same asset multiple times

* Refactor to support wmill s3 API

* nit

* parse_wmill_sdk_sql_assets refactor

* Detect datatable and ducklake usage

* nit order by

* Join with v2_job

* better UI

* add sequential id for cursor pagination

* useInfiniteQuery

* useScrollToBottom

* sql index

* claude code stash

* migration fixes

* Infinite scroll UI

* nit

* style nit

* runtime asset created at

* Asset filters

* fix usage kind filter

* also check runnable_path for jobs when filtering

* better filters

* avoid flickering

* debounced filters

* nit

* tooltips

* fix: update AssetUsage type to match new ListAssetsResponse structure

The ListAssetsResponse changed from an array to an object with an 'assets' property.
Updated the type extraction accordingly.

Co-authored-by: Diego Imbert <diegoimbert@users.noreply.github.com>

* sqlx prepare

* Delete .claude/hooks/.symlink-manifest

* unnecessary dep

* nit refactor

* nit comment

* nit naming

* CI fix attempt 1

* ee ref

* nit remove alerts

* nit

* chore: update ee-repo-ref to 138a4f5f868f3bded5bb7cb77b222b532c07e4af

This commit updates the EE repository reference after PR #395 was merged in windmill-ee-private.

Previous ee-repo-ref: 7d3a21d53066726e97dfea9f117373299bc9318c

New ee-repo-ref: 138a4f5f868f3bded5bb7cb77b222b532c07e4af

Automated by sync-ee-ref workflow.

---------

Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Diego Imbert <diegoimbert@users.noreply.github.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
 2026-02-02 10:09:40 +00:00