litellm_staging_04_04_2026

[krrish-berri-2]

Relevant issues

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

🆕 New Feature
🐛 Bug Fix
🧹 Refactoring
📖 Documentation
🚄 Infrastructure
✅ Test

Changes

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Apr 14, 2026 3:50pm

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
7 out of 8 committers have signed the CLA.

✅ nehaaprasad
✅ Dmitry-Kucher
✅ hunterchris
✅ jaxhend
✅ Sameerlite
✅ yuneng-berri
✅ kulia26
❌ d 🔹

---

d 🔹 seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[greptile-apps]

Greptile Summary

This is a focused staging bundle of five targeted bug fixes: a TypeError crash in Cache.get_cache_key when preset_cache_key was already present in kwargs; SigV4 signature failures in S3Logger when object keys contain spaces; false-positive MCP prefix detection for hyphenated non-MCP tool names; garbled Presidio PII output caused by applying anonymizer output offsets to the original input text; and an infinite login-redirect loop when nginx-ingress adds HttpOnly to the server-set token cookie. Each fix is accompanied by new unit tests.

Confidence Score: 5/5

• Safe to merge — all five bug fixes are correct, well-tested, and backward-compatible.
• No P0 or P1 issues found. Every change has matching unit tests. The single P2 finding (inline imports in one test method) is a style nit that does not affect correctness or CI.
• No files require special attention.

Important Files Changed

Filename	Overview
litellm/caching/caching.py	Correctly filters preset_cache_key from kwargs before forwarding to _set_preset_cache_key_in_kwargs to fix a TypeError: got multiple values crash when kwargs already contained that key.
litellm/integrations/s3_v2.py	Uses the requests-prepared URL (prepped.url) for all S3 PUT/GET calls so that percent-encoding in object-key path segments matches the SigV4 canonical request — fixes signature failures for keys with spaces or special characters.
litellm/proxy/_experimental/mcp_server/utils.py	Extends is_tool_name_prefixed to accept known_server_prefixes; when provided, only the substring before the first separator is matched against actual registered prefixes, eliminating false positives for non-MCP tools like text-to-speech.
litellm/proxy/guardrails/guardrail_hooks/presidio.py	Refactors anonymize_text into three helpers: _post_presidio_anonymize, _finalize_presidio_anonymize_simple, and _finalize_presidio_anonymize_numbered_tokens. Correctly fixes the position-reference bug by using analyzer positions (original text) instead of anonymizer item positions (output text) when building numbered PII tokens.
ui/litellm-dashboard/src/utils/cookieUtils.ts	Adds storeLoginToken (stores token in sessionStorage + JS cookie at /ui), updates getCookie to fall back to sessionStorage for the "token" key, and adds sessionStorage.removeItem("token") to clearTokenCookies. Correctly uses sessionStorage per CLAUDE.md guidance.
tests/test_litellm/integrations/test_s3_v2.py	New test verifies that S3 PUT uses a percent-encoded URL; contains inline imports inside the test method body (violates CLAUDE.md style guide).

Sequence Diagram

sequenceDiagram
    participant Browser
    participant Proxy as LiteLLM Proxy
    participant nginx

    Browser->>Proxy: POST /login/v2
    Proxy-->>nginx: 200 OK + Set-Cookie header
    nginx-->>Browser: 200 OK, Set-Cookie with HttpOnly added by nginx
    Note over Browser: Server cookie is HttpOnly - invisible to JS

    Browser->>Browser: storeLoginToken(data.token from JSON body)
    Note over Browser: Sets JS cookie at path /ui and writes to sessionStorage

    Browser->>Browser: getCookie("token")
    alt JS cookie found at /ui
        Browser-->>Browser: returns cookie value
    else cookie blocked
        Browser-->>Browser: returns sessionStorage fallback
    end

Loading

_{Reviews (5): Last reviewed commit: "Fix tests" | Re-trigger Greptile}

[greptile-apps]

Test placed in wrong directory

This file contains pure unit tests with no real network calls, so it belongs in tests/test_litellm/ (alongside the other new tests added by this PR) rather than tests/local_testing/. Tests under tests/local_testing/ are excluded from make test-unit, meaning this regression guard won't run in the standard CI pipeline.

Consider moving it to tests/test_litellm/caching/test_cache_preset_key.py.

[codspeed-hq]

Merging this PR will not alter performance

✅ 16 untouched benchmarks

---

_{Comparing litellm_oss_staging_04_04_2026 (69bf2bf) with main (e64d98f)}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[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
29203053	Triggered	Generic Password	972e42c	.circleci/config.yml	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.}