drift: close upstream interval 9028562..56ac317 (v1.35.0 -> v1.36.0) - #17
Merged
Conversation
9x NO_TARGET_CHANGE, 1x DEVIATION (DEV-SDK-001 for ToolShieldLLMSecurityAnalyzer). Adds TS regression port of the upstream MCP nested-schema preservation tests (issue #3955) as evidence for the schema-passthrough classification. Co-authored-by: openhands <openhands@all-hands.dev>
Ports upstream tests/sdk/mcp/test_mcp_nested_schema.py (issue #3955). The TS MCPToolDefinition passes inputSchema verbatim into LLM-facing payloads, so nested properties/required/descriptions survive; these tests pin that behavior against regressions. Co-authored-by: openhands <openhands@all-hands.dev>
Regenerates the Python projection oracle from the pinned checkout; all case results are identical to the previous pin. Co-authored-by: openhands <openhands@all-hands.dev>
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Closes upstream drift interval
9028562..56ac317(OpenHands/software-agent-sdk v1.35.0 -> v1.36.0), advancing the canonical pin to56ac31719f91059dc9b319dbc6bb79f17ef60cd7.Inventory generated with the repo's drift tooling (
drift:prepare); every first-parent commit x target unit received exactly one disposition before coding.drift:check --phase reviewpassed before any code change; close-phase validation passes at the new pin.Disposition table
Per-unit notes
openhands/sdk/security/; analyzers/risk scoring are intentionally not ported as active SDK behavior_load_auto_skills; TS AgentContext has no public-skills/marketplace auto-loading surfaceinputSchemaverbatim into LLM payloads, so the flattened-properties bug is structurally absent - regression port added as evidencePorted behavior changes
No PORT items this interval. The one behavioral surface touched (MCP tool schema construction, #4011) already satisfies its observable contract in TypeScript:
MCPToolDefinition.toOpenAiTool()/toResponsesTool()pass the raw MCPinputSchemathrough verbatim, preserving nested object properties, required arrays, enums, and descriptions.As evidence, ported the upstream regression suite
tests/sdk/mcp/test_mcp_nested_schema.py(issue #3955) intosrc/mcp/__tests__/mcp.test.ts: issue reproduction, Responses-facing preservation, three-level nesting, arrays of objects, and flat schemas - 7 new tests.Known pre-pin divergence noted in the review record (out of interval scope): upstream's summary/security_risk schema-field injection (predates initial pin, PR #1339) remains absent from the TS tool layer.
Test / parity results
npm test: 44 files, 349 tests passed (incl. 7 new MCP regression tests)npm run test:drift: 6/6 passednpm run typecheck+typecheck:drift: cleannpm run lint: cleannpm run build: success56ac317, all case results identical to previous pin;check-projection-parity.ts: 2/2 cases match exactlydrift:check --phase reviewand--phase close: validEvent/tool wire oracles remain follow-up work per docs/DRIFT_TOOLING.md ("differential oracles" section); those checks are unchanged from baseline and are not part of CI.