* fix: accept bunnative language in AI chat flow step validation
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: regenerate copilot flow schema from openflow spec
Run gen_openflow_schema.sh + minifiedOpenflowJson.sh instead of hand-patching. Also syncs three fields the checked-in generated files had drifted from since the last regen (reasoning_effort, reasoning_token_delta streaming event, aiagent tag).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: regenerate system prompts for bunnative openflow schema
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(ai-agent): support reasoning effort in AI agent workflow steps
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): round-trip native Anthropic thinking blocks and fix DeepSeek/Mistral reasoning
Address review: native Anthropic now captures the signed thinking block during streaming and replays it before tool_use across iterations (prevents a 400 on multi-turn tool use). DeepSeek 'off' sends thinking:{type:disabled} instead of the rejected reasoning_effort:none, and Mistral drops temperature when reasoning is on.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(ai-agent): move reasoning effort into the provider/model selector
Store reasoning_effort on ProviderConfig (next to the model) instead of a separate flow arg, and render the selector inside AIProviderPicker under the model dropdown. Add an explicit 'off' option on models that disable reasoning by omission (e.g. Claude), so reasoning can always be turned off from the UI.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* refactor(ai-agent): use DropdownV2 for reasoning effort, matching copilot chat
Replace the Select combobox with the same DropdownV2 action-menu the copilot chat reasoning selector uses. Each option carries an action instead of a bound value, so click selection is unambiguous and there is no typeahead/sentinel-value mismatch on the off/default entries.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(ai-agent): regenerate system prompts for ProviderConfig.reasoning_effort
Refresh system_prompts/auto-generated and cli skills.gen after adding reasoning_effort to the OpenFlow ProviderConfig schema (check-freshness).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): clear stale reasoning effort on model change; dedup bedrock reasoning folding
Address cubic review: (P1) the reasoning picker now clears the stored effort when the newly selected model doesn't accept it (e.g. carrying 'xhigh' from Opus onto a model that tops out at 'high'), not only when the model can't reason at all. (P3) the proxy's accumulate_reasoning_delta now delegates to the shared bedrock_stream_event_to_reasoning_delta so worker and proxy folding can't drift.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(ai-agent): stream reasoning summary and show a thinking affordance in flow chat
Add StreamingEvent::ReasoningTokenDelta, emitted from every worker reasoning path (Anthropic native thinking deltas, Bedrock, Gemini thought parts, OpenAI-compatible reasoning_content, OpenAI Responses reasoning_summary_text with summary:auto). The flow chat parses it and renders a collapsible 'Thinking' affordance on the assistant message (thinking tokens bill regardless of display, so surfacing the summary is billing-neutral).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): persist streamed reasoning onto the finished chat message
Reasoning isn't stored server-side, so the completion re-poll (which swaps temp messages for the persisted ones) was dropping the streamed thinking summary. Carry it onto the final assistant message so the 'Thought process' affordance survives the run.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(ai-agent): match flow-chat thinking box to the copilot chat reasoning UI
Replace the plain <details> thinking affordance with the same bordered, collapsible reasoning box the copilot chat uses (chevron + Brain/spinner + 'Thinking' header, markdown body, expand-while-streaming/collapse-on-answer).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): attribute streamed reasoning per turn by content; drop duplicated comment
Address review: the completion-poll carry-over now matches each temp assistant turn's thinking summary to its persisted message by content, so a multi-turn response (reasoning -> tool call -> final answer) no longer misattributes an earlier turn's thinking to the final answer or drops intermediate turns. Also removes a leftover duplicated comment block above the AIReasoningEffortPicker effect.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): address review round 2 (carry-over edges, off-token validity, aria, test)
cubic round 2: (1) reasoning carry-over now consumes temp turns in order verifying content, so identical/empty-content multi-turn responses attribute thinking correctly and reasoning-only turns aren't dropped; (2) the picker's stale-value check only accepts the off token when the model can actually disable reasoning; (3) add aria-expanded to the Thinking toggle; (4) add a test for the failed tool_result path.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): add bottom margin below the flow-chat thinking box
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): don't request OpenAI reasoning summary, matching the copilot chat
OpenAI gates reasoning summaries behind org verification, so requesting summary: auto would 400 for unverified orgs. The copilot chat requests effort only and never asks for a summary; align the worker with it (reasoning: { effort }) and drop the now-unreachable summary-delta parsing. OpenAI reasoning no longer streams a summary in flow chat (consistent with the copilot); Anthropic/Bedrock/Gemini/DeepSeek reasoning display is unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): scope reasoning carry-over to newly persisted messages
cubic round 3: matching by content over the full history could attach a new turn's reasoning to an older message with identical text. Restrict eligible targets to the messages just fetched for this response (via afterSeq), so historical turns are never touched.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): carry reasoning newest-first instead of gating on the final poll
cubic round 4: gating the carry-over on the final poll's filteredResponse dropped reasoning for messages already fetched by an earlier streaming poll (their id is excluded by afterSeq). Walk persisted newest-first and consume the newest matching pending summary, stopping once summaries run out. This response's turns are always at the end, so they claim their own reasoning (P1) before older history is reached (P2), regardless of which poll persisted them.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* refactor(ai-agent): drop flow-chat reasoning display, keep backend + effort picker
The chat-side thinking box relied on non-deterministically matching streamed (ephemeral) reasoning back onto persisted messages, which kept spawning edge cases. Remove the flow-chat display entirely (ChatMessage box, FlowChatManager carry-over/threading, parseStreamDeltas reasoning) and keep the sound backend: per-provider reasoning-effort requests, thinking-block round-trips for tool calls, and ReasoningTokenDelta streaming. A display can be built on top later, deterministically (e.g. once the stream carries the persisted message id).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai-agent): include reasoning_effort in default-config compare; document reasoning_token_delta
Codex/Pi nits: isSameAsStoredConfig now compares reasoning_effort so the 'use as personal default' toggle reflects effort-only changes; openflow streaming-events doc lists the reasoning_token_delta event (regenerated auto prompts).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(ai): add Azure AI Foundry as a native AI provider
Adds `azure_foundry` as a new AIProvider variant wired through the AI
chat (copilot) and AI agent flow steps. Foundry's chat completions API
is OpenAI-compatible and uses Azure conventions (api-key header, Azure
URL building), so it reuses the existing OpenAI-compatible query builder
and proxy path via the shared `is_azure` helper (renamed from
`is_azure_openai`).
Backend (windmill-ai):
- New `AzureFoundry` enum variant (serde `azure_foundry`)
- `get_base_url` requires a resource base URL (like Azure OpenAI / Custom)
- `is_azure()` covers Azure OpenAI + Foundry (api-key auth, Azure URL)
- Added to OpenAI-compatible proxy support and HttpForward proxy mode
- New proxy URL unit test
Frontend (copilot):
- New provider entry, completion config, model-token handling, streamed
usage tracking, and reasoning registry (all model-id-gated, so a no-op
for Foundry's non-OpenAI catalog)
- Treated as a chat-completions provider, not the OpenAI Responses API
OpenAPI:
- `azure_foundry` added to AIProvider (openapi.yaml) and AIProviderKind
(openflow.openapi.yaml); regenerated CLI guidance
Note: the `azure_foundry` resource type (base_url + optional api_key) is
hub-managed and must be published to the Windmill Hub separately.
Fixes WIN-2122
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ai): add azure_foundry to copilot flow Zod provider enum
The tracked copilot flow schema (openFlowZod.gen.ts and its openFlow.json
source) still carried the old AIProvider enum, so validateFlowModules /
validateSpecialFlowModule rejected AI-generated flow edits that create or
update an aiagent module with provider kind "azure_foundry" before they
could be saved. Add the value to both (preserving the generated single-line
format) and a regression test over the flow-module validation path.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(ai): lead provider list with OpenAI, Anthropic, Google AI
Reorder AI_PROVIDERS so the three primary direct providers come first. The
AIProviderPicker renders the first three entries as quick-access buttons, so
these become the defaults (previously OpenAI, Azure OpenAI, Azure Foundry);
Azure OpenAI / Azure Foundry stay adjacent right after. No logic depends on
provider order (only per-provider defaultModels[0] is read).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* 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
* feat(flows): early stop can include the stopping step's result in the raised error
When a step uses Early Stop with "Raise an error message if stopped", the
flow result was entirely replaced with a static error object
({"error": {"name": "EarlyStopError", "message": "..."}}), discarding the
stopping step's own output. This made it impossible to stop+fail a flow
while preserving the data the step produced (e.g. an API that returns
HTTP 200 with a userErrors payload).
Add an opt-in `error_include_result` flag on StopAfterIf. When enabled on
the raise-error path, the raised payload becomes
{"error": {...}, "result": <step result>} instead of dropping the result.
Default is false, so existing behavior is unchanged. The option is threaded
through the worker's stop-after-if handling (including stop_after_all_iters_if
for loops/branchall) and exposed in the flow editor's Early Stop panel.
Fixes WIN-2012
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(flows): cover early-stop error_include_result payload shaping
Add a regression test asserting that a step using Early Stop with a raised
error message and error_include_result=true fails the flow while preserving
the step output as {"error": {..}, "result": <step result>}, and that with
the flag off the result is the bare {"error": {..}} object.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* refactor(flows): nest early-stop step result inside the error object
Embed the stopping step's result under `error.result` rather than as a
top-level sibling of `error`. This keeps the flow result shape as
`{ "error": { .. } }` — identical to a normal error — so consumers that
key off the top-level shape (single `error` key) keep working, while the
data is still preserved for those that look inside the error object.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(flows): always include the stopping step's result in early-stop errors
Drop the opt-in `error_include_result` gate. Since the step result is nested
inside the error object (`error.result`), the top-level result shape stays
`{ "error": .. }` — identical to a normal error — so consumers that detect or
parse failures by the top-level shape are unaffected. Gating it added schema
surface, plumbing, and a UI toggle for no real compatibility benefit.
Now, whenever a step early-stops with a raised error message, the flow fails
and the raised error embeds the stopping step's own result under
`error.result` (aggregated iteration results for loops/branchall). This
reverts the `StopAfterIf.error_include_result` field, its threading, the
OpenAPI/generated-client surface, and the editor toggle; the "Raise an error
message" tooltip now notes that the step result is included.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(flows): gate early-stop result inclusion behind opt-in flag
Re-introduce the per-step `error_include_result` flag (default off) instead
of always embedding the step result. Although nesting the result under
`error.result` keeps the result *shape* backward-compatible, it does not
address data exposure: a failed flow's result is propagated to synchronous
webhook callers, the flow's failure module, and the workspace/global error
handler (commonly a Slack/email/outbound-webhook notifier). Always including
the step output would surface previously-redacted intermediate data to all of
those sinks for every existing error-stop flow.
Gating keeps the existing behavior (bare `{ "error": .. }`) as the default and
only embeds `error.result` when the flow author explicitly opts in, matching
the original issue's intent.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(flows): omit error_include_result when false; refresh generated prompts
- Add `skip_serializing_if = "is_false"` to `StopAfterIf.error_include_result`
so serialized flows are byte-identical when the flag is off. Fixes the
`flowmodule_serde` round-trip test (cargo_test) and avoids churn on existing
flows.
- Regenerate `system_prompts/auto-generated/` and `cli/src/guidance/skills.gen.ts`
for the new OpenFlow `error_include_result` property. Fixes check-freshness.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(flows): cover error_include_result for the loop "stop after all iters" path
Add a regression test for the stop_after_all_iters_if branch, where `nresult`
already holds the aggregated iteration results — confirming `error.result`
carries each iteration's output (distinct from the per-step fallback path).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Ruben Fiszel
hugocasa