feat(mcp): pass session_id through MCP tool calls

[erichare]

Summary

• Add optional session_id parameter to flows exposed via the MCP interface so MCP clients can persist chat history across tool calls.
• json_schema_from_flow now advertises session_id in each tool's input schema (optional, not in required).
• handle_call_tool uses the client-provided session_id when present and falls back to a generated UUID otherwise (preserves prior behavior).

Reported by Alejandro — MCP clients had no way to maintain chat history because the server unconditionally generated a fresh conversation id per call, so memory components could never see prior turns.

Files Changed

• src/backend/base/langflow/helpers/flow.py — inject optional session_id into the generated JSON schema.
• src/backend/base/langflow/api/v1/mcp_utils.py — extract session_id from arguments; fall back to UUID when absent/blank.
• src/backend/tests/unit/api/v1/test_mcp_utils.py — 5 new unit tests covering both pass-through and fallback paths plus the schema additions.

Test plan

• uv run pytest src/backend/tests/unit/api/v1/test_mcp_utils.py — 19/19 pass (5 new).
• uv run ruff check / uv run ruff format --check on touched files — clean.
• Manual: call an MCP tool twice with the same session_id; second call sees prior turn in chat history.
• Manual: call without session_id; flow still runs as before (fresh session each call).
• Manual: tools/list against the MCP server shows session_id in inputSchema.properties and not in required.

Backward compatibility

Existing MCP clients that do not send session_id are unaffected — they continue to get an auto-generated UUID per call, matching today's behavior.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 856b1036-1f6c-4b1b-8102-03145825f99a

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/mcp-session-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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 54.70%. Comparing base (e21f588) to head (765d63e).

Additional details and impacted files

@@                Coverage Diff                 @@
##           release-1.10.0   #13124      +/-   ##
==================================================
- Coverage           54.71%   54.70%   -0.01%     
==================================================
  Files                2145     2145              
  Lines              200199   200142      -57     
  Branches            30199    28490    -1709     
==================================================
- Hits               109537   109491      -46     
+ Misses              89481    89469      -12     
- Partials             1181     1182       +1     

Flag	Coverage Δ
backend	60.14% <100.00%> (+0.20%)	⬆️
frontend	54.66% <ø> (-0.06%)	⬇️
lfx	49.30% <ø> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/backend/base/langflow/api/v1/mcp_utils.py	65.56% <100.00%> (+12.08%)	⬆️
src/backend/base/langflow/helpers/flow.py	60.83% <100.00%> (+7.47%)	⬆️

... and 74 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Frontend Unit Test Coverage Report

Coverage Summary

Lines	Statements	Branches	Functions
	38.4% (47871/124659)	67.9% (6542/9634)	38.28% (1100/2873)

Unit Test Results

Tests	Skipped	Failures	Errors	Time
4302	0 💤	0 ❌	0 🔥	11m 20s ⏱️