Skip to content

πŸ›‘οΈ Sentinel: [MEDIUM] Fix internal error leakage via JSON-RPC responses#239

Open
davidruzicka wants to merge 1 commit into
mainfrom
sentinel/fix-json-rpc-error-leakage-12616939701943940330
Open

πŸ›‘οΈ Sentinel: [MEDIUM] Fix internal error leakage via JSON-RPC responses#239
davidruzicka wants to merge 1 commit into
mainfrom
sentinel/fix-json-rpc-error-leakage-12616939701943940330

Conversation

@davidruzicka
Copy link
Copy Markdown
Owner

🚨 Severity: MEDIUM
πŸ’‘ Vulnerability: The MCP server's handleOtherRequest method was catching exceptions and returning the raw error message directly to the client in JSON-RPC responses for methods like prompts/get, resources/read, and completion/complete.
🎯 Impact: Exposing raw internal errors to clients can leak sensitive information such as file paths, database structures, or underlying software configurations, aiding attackers in reconnaissance.
πŸ”§ Fix: I updated the catch blocks for these handlers to generate a correlation ID, log the raw error securely server-side, and return a sanitized, generic message to the client using this.formatErrorForClient(error, correlationId).
βœ… Verification: Verified by running the full test suite (npm run test), which asserts that Validation error: responses include the dynamic correlation ID and do not expose stack traces.


PR created automatically by Jules for task 12616939701943940330 started by @davidruzicka

This commit fixes a vulnerability where `handleOtherRequest` in the MCP server would leak raw internal error details (like file paths or stack traces) directly to the client in JSON-RPC responses for methods like `prompts/get`, `resources/read`, and `completion/complete`.

It resolves the issue by generating a unique correlation ID, securely logging the error server-side with that ID, and sending a sanitized, generic error message to the client using `this.formatErrorForClient(error, correlationId)`. This aligns these handlers with the secure error handling practice established in `.jules/sentinel.md`.

Co-authored-by: davidruzicka <14172985+davidruzicka@users.noreply.github.com>
@google-labs-jules
Copy link
Copy Markdown
Contributor

πŸ‘‹ Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a πŸ‘€ emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.


For security, I will only act on instructions from the user who triggered this task.

@chatgpt-codex-connector
Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

@codecov
Copy link
Copy Markdown

codecov Bot commented Apr 4, 2026

Codecov Report

βœ… All modified and coverable lines are covered by tests.
βœ… All tests successful. No failed tests found.

πŸ“’ Thoughts on this report? Let us know!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant