added support for metadata

[shivamrawat1]

Relevant issues

ChatGPT Apps and MCP Apps depend on Resource metadata (_meta), but the LiteLLM proxy’s MCP read_resource handler was not including metadata in its responses.

When normalizing TextResourceContents and BlobResourceContents to ReadResourceContents, the code built new objects with only content and mime_type, so upstream meta was dropped.

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/test_litellm/ directory, Adding at least 1 test is a hard requirement - see details
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem
• I have requested a Greptile review by commenting @greptileai and received a Confidence Score of at least 4/5 before requesting a maintainer review

Delays in PR merge?

If you're seeing a delay in your PR being merged, ping the LiteLLM Team on Slack (#pr-review).

CI (LiteLLM team)

CI status guideline:

• 50-55 passing tests: main is stable with minor issues.
• 45-49 passing tests: acceptable but needs attention
• <= 40 passing tests: unstable; be careful with your merges and assess the risk.

• Branch creation CI run
  Link:

• CI run for the last commit
  Link:

• Merge / cherry-pick CI run
  Links:

Type

🐛 Bug Fix
✅ Test

Changes

Extracted normalization into _normalize_resource_contents() and forward meta from the source content.
Handle Pydantic models that store meta under _meta by falling back to model_dump() when getattr(content, "meta", None) is None.
Bumped MCP dependency to 1.26.0, requirements.txt, .circleci/config.yml, and .circleci/requirements.txt.
Added tests to ensure metadata is preserved and behavior remains correct when meta is absent.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Apr 4, 2026 6:38pm

[shivamrawat1]

@greptile review this

[codspeed-hq]

Merging this PR will not alter performance

✅ 16 untouched benchmarks

---

_{Comparing litellm_metadata_for_mcp (12fa28f) with main (c898edb)}

[greptile-apps]

Greptile Summary

This PR fixes a bug in the LiteLLM proxy's MCP read_resource handler where resource metadata (_meta) was silently dropped during normalization of TextResourceContents and BlobResourceContents to ReadResourceContents.

Key changes:

• Extracted the inline normalization loop in mcp_read_resource into a dedicated _normalize_resource_contents() helper function
• Added meta forwarding for both TextResourceContents and BlobResourceContents, using getattr(content, \"meta\", None) with a model_dump() fallback that also checks _meta for older MCP versions
• Fixed a pre-existing omission: BlobResourceContents.mimeType is now forwarded (it was previously hardcoded to None)
• Bumped the mcp dependency to exactly 1.26.0 in pyproject.toml and requirements.txt (consistent with the project's optional-dep pinning convention)
• Removed unused Optional import and cleaned up test imports
• Added four unit tests covering: text metadata, blob metadata, empty-dict metadata, and the no-metadata backward-compatibility case

All previously raised concerns have been addressed.

Confidence Score: 5/5

Safe to merge — the metadata-forwarding fix is correct, all previously raised concerns have been addressed, and the new tests cover the critical paths.

All P1 concerns from prior review rounds are resolved: BlobResourceContents.mimeType is now forwarded, empty-dict meta is handled with is None checks, blob coverage was added, and pyproject.toml/requirements.txt have been bumped to mcp 1.26.0. The only remaining stylistic item (unparameterized list annotation) was already flagged and is P2. No new P0 or P1 issues found.

No files require special attention.

Important Files Changed

Filename	Overview
litellm/proxy/_experimental/mcp_server/server.py	Extracted inline normalization loop into _normalize_resource_contents(), adding meta forwarding and fixing the pre-existing BlobResourceContents.mimeType=None bug.
tests/test_litellm/proxy/_experimental/mcp_server/test_mcp_server.py	Added four new unit tests for _normalize_resource_contents (text meta, blob meta, empty-dict meta, no-meta); removed unused imports; existing tests unmodified.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[mcp_read_resource handler] -->|calls| B[_normalize_resource_contents]
    B --> C{for each content item}
    C --> D[getattr content.meta]
    D -->|None| E{hasattr model_dump?}
    D -->|value| H[use meta value]
    E -->|yes| F[content.model_dump]
    E -->|no| G[meta = None]
    F --> F1{d.get meta}
    F1 -->|value| H
    F1 -->|None| F2{d.get _meta}
    F2 -->|value| H
    F2 -->|None| G
    H --> I{isinstance?}
    G --> I
    I -->|TextResourceContents| J[ReadResourceContents content=text mime_type=mimeType meta=meta]
    I -->|BlobResourceContents| K[ReadResourceContents content=blob mime_type=mimeType meta=meta]
    I -->|other| L[silently skipped]
    J --> M[append to normalized]
    K --> M
    M --> N[return List of ReadResourceContents]

Loading

_{Greploops — Automatically fix all review issues by running /greploops in Claude Code. It iterates: fix, push, re-review, repeat until 5/5 confidence.
Use the Greptile plugin for Claude Code to query reviews, search comments, and manage custom context directly from your terminal.}

_{Reviews (6): Last reviewed commit: "Merge branch 'main' into litellm_metadat..." | Re-trigger Greptile}

[shivamrawat1]

@greptile re-review this now that all comments are resolved.

[shivamrawat1]

@greptile review given that we can ignore Unknown content types silently dropped issue and that description is updated to reflect that we did not change it in py.project

[shivamrawat1]

@greptile re-review this with all commits and give score out of 5.

[gitguardian]

⚠️ GitGuardian has uncovered 1 secret following the scan of your pull request.

Please consider investigating the findings and remediating the incidents. Failure to do so may lead to compromising the associated services or software components.

🔎 Detected hardcoded secret in your pull request

GitGuardian id	GitGuardian status	Secret	Commit	Filename
29375658	Triggered	JSON Web Token	12fa28f	tests/test_litellm/proxy/auth/test_handle_jwt.py	View secret

🛠 Guidelines to remediate hardcoded secrets

1. Understand the implications of revoking this secret by investigating where it is used in your code.
2. Replace and store your secret safely. Learn here the best practices.
3. Revoke and rotate this secret.
4. If possible, rewrite git history. Rewriting git history is not a trivial act. You might completely break other contributing developers' workflow and you risk accidentally deleting legitimate data.

To avoid such incidents in the future consider

• following these best practices for managing and storing secrets including API keys and other credentials
• install secret detection on pre-commit to catch secret before it leaves your machine and ease remediation.

---

^{🦉 GitGuardian detects secrets in your source code to help developers and security teams secure the modern development process. You are seeing this because you or someone else with access to this repository has authorized GitGuardian to scan your pull request.}

[shivamrawat1]

@greptile review this. Note that pyproject is moved to 1.26.0