feat: add observability metadata to tools and guardrails #384
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add metadata to tools and guardrails for observability span detection. Changes: - escalation_tool.py: Add metadata (tool_type, display_name, channel_type, assignee) - process_tool.py: Add metadata (tool_type, display_name, folder_path) - guardrail_nodes.py: Wrap in RunnableCallable with metadata 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
RunnableCallable passes metadata as kwargs which breaks node invocation. Use __metadata__ attribute on function instead, which callbacks can read. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
Can be inferred from guardrail_name presence. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
- Add test_process_tool.py for process tool metadata - Add test_escalation_tool.py for escalation tool metadata - Add TestGuardrailNodeMetadata to guardrail node tests 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
- Use getattr() for __metadata__ access (avoids type:ignore) - Add assert metadata is not None before indexing 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
cristipufu
requested review from
andreitava-uip,
cotovanu-cristian and
valentinabojan
andreitava-uip
approved these changes
Dec 31, 2025
JosephMar
approved these changes
Dec 31, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
Add metadata to tools and guardrails for observability span detection.
Changes:
escalation_tool.py: Add metadata (tool_type,display_name,channel_type,assignee)process_tool.py: Add metadata (tool_type,display_name,folder_path)guardrail_nodes.py: Attach__metadata__attribute (guardrail_name,guardrail_scope,guardrail_stage,tool_name)Metadata usage
Core (span detection):
tool_type,display_name→ identify interruptible tool type and span nameSpan attributes (temporal parity):
channel_type,assignee,folder_path→ additional span attributesWhy
tool_namein guardrail metadata?For TOOL scope guardrails, correlates guardrail span with parent tool span for proper hierarchy.
Why
__metadata__for nodes butmetadatafor tools?StructuredToolhas built-inmetadataparam. Guardrail nodes are plain async functions—need manual__metadata__attribute that callbacks read viagetattr().Why
Metadata is the cleanest approach: it's the standard LangChain pattern, flows automatically to callbacks, and requires no parsing.
tool_typeis explicit because Python SDK tools share the sameStructuredToolbase class—unlike temporal's polymorphicGetToolType(), we can't derive type from class hierarchy.Removes name-based hacks in
uipath-agents-pythoncallback: