* feat: add Pydantic BaseModel and dataclass support for Python type inference
- Add AST-based detection of Pydantic BaseModel inheritance patterns
- Add AST-based detection of @dataclass decorator (all variants)
- Implement recursive field schema extraction with type inference
- Add thread-safe stack-based module storage for nested parsing
- Add RAII cleanup guard to ensure memory safety on all code paths
- Add security limits: 200 fields max, 10 recursion levels max
- Add comprehensive test coverage: 3 new tests for Pydantic/dataclass
- Maintain 100% backward compatibility with existing type system
This enables ML/AI practitioners to use Pydantic models as function
parameters with automatic UI generation from model schemas.
Implementation highlights:
- Zero code execution: Pure AST analysis for safety
- Thread-safe: Stack-based storage prevents race conditions
- Memory-safe: RAII pattern guarantees cleanup
- Security-hardened: Field count and recursion depth limits
- Performance-optimized: Depth-limited recursion, lazy parsing
Test results: All 12 tests passing (9 existing + 3 new)
Closes#4700🤖 Generated with Claude Code (https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
* fix: improve Pydantic/dataclass parser robustness and error handling
This commit addresses critical bugs and improves error handling in the
Python parser for Pydantic BaseModel and dataclass support.
## Critical Fixes
1. **Thread-local storage RAII pattern**: Fixed bug where parse failures
could leave the module stack in an inconsistent state. Now uses proper
functional composition with .ok().map() to ensure cleanup always happens.
2. **Recursion depth warnings**: Added explicit warning messages when the
recursion depth limit (10 levels) is reached during type extraction.
Made the limit a named constant for clarity.
3. **Unsupported type warnings**: Added informative warning messages for
unsupported type annotations (Union types and forward references) to
help users understand why their types aren't being inferred.
## Improvements
- Added 10 comprehensive test cases covering:
- Empty Pydantic models
- List[T] and Optional[T] types
- Dataclass with decorator arguments
- Dict types
- Regular classes (non-model types)
- Invalid syntax handling
- Datetime fields
- Multiple model definitions
- Nested models
- All 21 tests pass successfully
## Testing
Verified that:
- Parser handles malformed code gracefully
- RAII cleanup works correctly with early returns
- Warning messages are clear and actionable
- No memory leaks or panics
Closes#4700
* refactor: Separate Pydantic/dataclass code into dedicated module. Created src/pydantic_parser.rs with thread-local storage, model detection, and type extraction logic. Moved 12 Pydantic tests to tests/pydantic_tests.rs and removed duplicate code from lib.rs. All 21 tests passing.
* opti and publish
---------
Co-authored-by: Devdatta Talele <devtalele0@gmail.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
* feat: implement quiet mode to filter verbose logs
Add QUIET_MODE environment variable (set to "true" or "1" to enable) that:
- Filters logs with the "windmill_verbose" target via tracing_init
- Increases frequency of periodic logs by 10x when enabled
- Marks per-job worker logs as verbose so they are suppressed in quiet mode
Key changes:
- Add QUIET_MODE lazy_static and VERBOSE_TARGET constant to tracing_init.rs
- Create create_targets_filter() helper that filters windmill_verbose target
- Mark periodic "still running" and "memory snapshot" logs with VERBOSE_TARGET
- Mark per-job "fetched job", "started handling", "job finished" logs as verbose
- Increase memory snapshot interval from every 10 ticks to every 100 ticks in quiet mode
- Increase "still running" log interval from every tick to every 10 ticks in quiet mode
- Add quiet mode notification in job logs
Closes#7477🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* Update tracing_init.rs
* Update worker.rs
* Remove target from tracing info logs
---------
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
Co-authored-by: Ruben Fiszel <ruben@windmill.dev>
* use tini with unshare to preserve signals (e.g oom)
* fix ci
* ci as nsjail
* simplify
* fix flaky go integration test
---------
Co-authored-by: Ruben Fiszel <ruben@windmill.dev>
When updating rows in the Database Manager with MariaDB tables containing
DATE columns with non-null values, the update query was failing silently
because the date-only format (YYYY-MM-DD) was not recognized by the date
parsing regex. This caused the WHERE clause to use '0000-00-00' instead
of the actual date value. Added a new regex pattern to handle the date-only format that MariaDB
returns for DATE columns, ensuring proper date comparison in UPDATE
queries.
Co-authored-by: Diego Imbert <70353967+diegoimbert@users.noreply.github.com>
* fix: add CORS headers to static assets for iframe context sharing
Add Cross-Origin-Opener-Policy, Cross-Origin-Embedder-Policy, and
Cross-Origin-Resource-Policy headers to static assets served by the backend
to match the vite preview config. This enables the TypeScript worker to work
correctly when ui_builder is loaded in an iframe.
Fixes#7453
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* fix: add CORS headers only for /apps_raw paths and force reload on navigation
- Only add Cross-Origin-Opener-Policy, Cross-Origin-Embedder-Policy, and
Cross-Origin-Resource-Policy headers for /apps_raw/ paths (not all static assets)
- Add frontend navigation handler to force page reload when navigating from
non-apps_raw paths to /apps_raw/add or /apps_raw/edit
- This ensures the TypeScript worker works correctly while avoiding CORS issues
for external iframe embeds in other parts of the app
Co-authored-by: Ruben Fiszel <rubenfiszel@users.noreply.github.com>
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: also reload when navigating from /apps/get_raw/ to apps_raw editor
When viewing a raw app at /apps/get_raw/, the cross-origin isolation
headers are not present. Force a page reload when navigating from
this path to /apps_raw/add or /apps_raw/edit to ensure the headers
are fetched from the server.
Co-authored-by: Ruben Fiszel <rubenfiszel@users.noreply.github.com>
* Update +layout.svelte
---------
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Ruben Fiszel <rubenfiszel@users.noreply.github.com>
Co-authored-by: Ruben Fiszel <ruben@windmill.dev>
* fix frontend
Signed-off-by: pyranota <pyra@duck.com>
* chore: update ee-repo-ref to 5c2a8854e7ff014063a69dd8f7829a935129c31e
This commit updates the EE repository reference after PR #370 was merged in windmill-ee-private.
Previous ee-repo-ref: 8c75fc5b1372a790dc74b374cd72f9caf51a4595
New ee-repo-ref: 5c2a8854e7ff014063a69dd8f7829a935129c31e
Automated by sync-ee-ref workflow.
---------
Signed-off-by: pyranota <pyra@duck.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* refactor(ai): use responses API for OpenAI/Azure, create 'other' provider for completion endpoint
- Created new 'other.rs' provider using the OpenAI-compatible completion endpoint
- Refactored 'openai.rs' to use the Responses API for both text and image output
- Updated query_builder to route OpenAI/AzureOpenAI to OpenAIQueryBuilder
- All other providers (Mistral, DeepSeek, Groq, etc.) now use OtherQueryBuilder
- Updated OpenRouter to delegate to OtherQueryBuilder instead of OpenAIQueryBuilder
This prepares the codebase for adding websearch tool support using the Responses API.
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* feat(ai): add websearch as AI agent tool type
- Added WebsearchToolValue to ToolValue enum in flows.rs
- Updated all pattern matches to handle websearch tool type
- Added has_websearch parameter to run_agent function
- Websearch tools don't require additional configuration
This prepares the backend for provider-specific websearch implementations.
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* feat(frontend): add websearch as tool option in AI agent UI
- Added WebsearchTool type and utility functions to agentToolUtils.ts
- Added "Web Search" option in tool insertion menu (InsertModuleInner)
- Updated NewAIToolNode to handle pickWebsearchTool event
- Updated AIToolNode to render websearch tools properly
- Updated FlowModuleSchemaMap to create websearch tools
Users can now add websearch tools to AI agents through the UI.
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* feat(ai): implement websearch for OpenAI provider
- Added has_websearch parameter to BuildRequestArgs
- OpenAI provider now adds web_search tool to requests when enabled
- Uses OpenAI Responses API web_search tool type
- Websearch tool is added before other custom tools in the request
Implements websearch functionality for OpenAI and Azure OpenAI providers.
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* feat(ai): implement websearch for Anthropic provider
- Created anthropic.rs provider with native Anthropic API format
- Added web_search tool to Anthropic requests when enabled
- Anthropic uses /messages endpoint with x-api-key authentication
- Updated query_builder to route Anthropic to new provider
Implements websearch functionality for Anthropic Claude models.
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* feat(ai): implement websearch for Gemini/GoogleAI provider
- GoogleAI now uses completion endpoint (other.rs) for text instead of responses API
- Added Google Search grounding when websearch is enabled
- Uses google_search_retrieval tool in request when has_websearch is true
- Updated parse methods to use OtherQueryBuilder for completion endpoint
Implements websearch functionality for Google Gemini models.
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* fix frontend
* fix anthropic and openai
* better for gemini
* structured output
* cleaning
* fix validate tool
* fixes
* cleaning
* cleaning
* fix for openai
* no responses api for azure
* fixes
* fix
* add tests for ai agent
* avoid panic
* better tests
* test user images
* fix tool choice
* always use streaming backend side
* big cleaning
* show annotations plus agent action for open ai websearch use
* show annotations plus agent action for anthropic websearch use
* show annotations plus agent action for google websearch use
* nit forntend
* rm
* fix
* add test for image ouptut
* fix for azure
* add in openflow
* fix
* fix
* nit tests
* fixes
---------
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
Co-authored-by: centdix <farhadg110@gmail.com>
* handle messages array for ai agent
* better
* nit
* make tool_calls and tool_call_id nullable
* fix empty json behavior
* nits
* cleaning
* feat(backend): replace messages/messages_context_length with history oneOf field
Replace the separate 'messages' array and 'messages_context_length' fields with
a single 'history' field that uses a oneOf discriminator.
The 'history' field can be either:
- 'auto' mode: automatically manages conversation history with memory, takes a
'context_length' number parameter
- 'manual' mode: bypasses memory and uses explicitly provided messages array
Backward compatibility is maintained: if 'messages_context_length' is provided
in the old schema format, it is automatically converted to 'auto' mode with the
specified context_length.
Co-authored-by: centdix <centdix@users.noreply.github.com>
* feat(frontend): replace messages/messages_context_length with history oneOf field
Replace the separate 'messages' array and 'messages_context_length' fields with
a single 'history' field in the AI agent schema.
The 'history' field uses a oneOf discriminator with two modes:
- 'auto': { mode: 'auto', context_length: number } - automatically manages
conversation history with memory
- 'manual': { mode: 'manual', messages: array } - bypasses memory and uses
explicitly provided messages
The schema includes comprehensive descriptions for each mode explaining the
behavior. The order array has been updated to include 'history' in place of
the old 'messages_context_length' and 'messages' fields.
Co-authored-by: centdix <centdix@users.noreply.github.com>
* fix(frontend): add support for 'mode' discriminator in oneOf rendering
Update ArgInput.svelte to properly handle oneOf schemas that use 'mode' as the
discriminator field, in addition to the existing 'kind' and 'label' support.
Changes:
- Updated tagKey derivation to check for 'mode' first, then 'kind', then 'label'
- Added 'mode' to the onOneOfChange function to track mode changes
- Added 'mode' to the list of keys excluded from enum validation
- Added 'mode' to hiddenArgs to prevent it from being shown in the form
- Added title fields to the history oneOf variants in flowInfers.ts
This allows the AI agent's history field to properly render with toggle buttons
for 'auto' and 'manual' modes.
Co-authored-by: centdix <centdix@users.noreply.github.com>
* fixes
* frontend fix
* nit
* cleaning
* cleaning
* better
* reword
* reword
---------
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: centdix <centdix@users.noreply.github.com>
* Fix MS SQL Numeric rounding
* Fix result collection to avoid JSON parsing
* don't reparse json through sqlx, pass raw string to postgres
* Revert "don't reparse json through sqlx, pass raw string to postgres"
This reverts commit 355691fa7d.
* mistake
---------
Co-authored-by: Ruben Fiszel <ruben@windmill.dev>
* fix: Migrate to astral-tokio-tar to address CVE-2025-62518
Migrates from the abandoned tokio-tar library to the actively maintained astral-tokio-tar to address CVE-2025-62518 (TARmageddon), a high-severity RCE vulnerability (CVSS 8.1). The vulnerability involves a boundary-parsing bug in PAX/ustar header handling that enables file smuggling attacks via malicious TAR archives.
This is a drop-in replacement requiring only dependency updates in Cargo.toml files. The astral-tokio-tar library uses the same tokio_tar module name, so no source code changes are needed. All references to the vulnerable tokio-tar v0.3.1 have been removed from Cargo.lock and replaced with the patched astral-tokio-tar v0.5.6.
Related to previous PR #6943 which was closed due to CLA issues. Security disclosure available at https://edera.dev/stories/tarmageddon and patch details at https://github.com/astral-sh/tokio-tar/releases/tag/v0.5.6
* update
---------
Co-authored-by: Devdatta Talele <devtalele0@gmail.com>
Co-authored-by: Devdatta Talele <50290838+devdattatalele@users.noreply.github.com>
Diego Imbert
centdix
hugocasa
Pyra
Alexander Petric
claude[bot]
Devdatta Talele
Ruben Fiszel
Aananth K