fix(scripts): end make check with a ran/skipped summary and verdict

[mateo-berri]

TLDR

Problem this solves:

• make check could exit 0 printing nothing, leaving a zero-byte log
• staged files matching no check looked identical to a crashed run
• devs re-ran it, then read the script, to learn nothing ran

How it solves it:

• every run now ends naming each check as ran or skipped
• when nothing matches, a NOTE lists the in-scope files and says the pass is a no-op
• a final check: PASS or check: FAIL line mirrors the exit code
• early informational exits (nothing to check, no merge base) print the verdict too

User Flow

Before: a contributor whose staged files match no lint check gets a silent exit 0 and a zero-byte log, indistinguishable from the script dying early

1. They stage a cost-map JSON and a tests/test_litellm test file with git add, then run make check
2. The terminal prints only "check: logging full output to .git/pre_commit_lint.log" and "check: full log: ..." with nothing in between, and exits 0
3. They open the log file and find it empty, so a clean pass looks exactly like a run that silently broke
4. They re-run make check a few times, then have to read scripts/pre_commit_lint.sh itself to learn that nothing was ever supposed to run

After: the same run ends with a per-check ran/skipped summary, the reason nothing ran, and an explicit verdict

1. They stage a cost-map JSON and a tests/test_litellm test file with git add, then run make check
2. The terminal prints the same first log line, then a "check: summary" block naming all four checks as skipped with the reason, like "no litellm/ Python files in scope"
3. A NOTE lists the staged files, says no gating lint check matches them, and that the pass is a no-op, not a lint verdict
4. The run ends with "check: PASS" ("check: FAIL" on a red run), and the log file holds that same summary instead of being empty

Relevant issues

Linear ticket

Pre-Submission checklist

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

• I have added meaningful tests
• My PR passes all CI/CD checks (e.g., lint, format, unit tests)
• My PR's scope is as isolated as possible; it only solves 1 specific problem
• I have received a Greptile Confidence Score of at least 4/5 before requesting a maintainer review (Greptile reviews automatically once the PR is opened; only comment @greptileai to re-request a review after pushing changes)

Delays in PR merge?

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

Screenshots / Proof of Fix

Scenario from the bug report: a file no gating check covers (a tests/test_litellm Python file) is the only thing staged, in a bootstrapped worktree

Before, running the script as of the merge base 9d069f2:

$ git show origin/litellm_internal_staging:scripts/pre_commit_lint.sh > /tmp/base_pre_commit_lint.sh && chmod +x /tmp/base_pre_commit_lint.sh
$ git add tests/test_litellm/tmp_proof_scratch.py
$ /tmp/base_pre_commit_lint.sh; echo "exit=$?"
check: logging full output to /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log
check: full log: /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log
exit=0
$ wc -c /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log
       0 /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log

After, same staged file at cfbd431:

$ scripts/pre_commit_lint.sh; echo "exit=$?"
check: logging full output to /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log
check: summary
    skipped: Python lint (make lint) (no litellm/ Python files in scope)
    skipped: tests/e2e checks (basedpyright + raw HTTP client ban) (no tests/e2e Python files in scope)
    skipped: dashboard lint (prettier + eslint + lint budgets) (no dashboard files in scope)
    skipped: dashboard API-type sync (npm run gen:api) (no litellm/proxy, litellm/types, or generator files in scope)
check: NOTE - no gating lint check matches the files in scope, so nothing ran:
    tests/test_litellm/tmp_proof_scratch.py
  A pass here is a no-op, not a lint verdict.
check: PASS
check: full log: /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log
exit=0
$ wc -c /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log
     584 /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-check-summary/pre_commit_lint.log

Early-exit verdict at 3fbe40c, in a clean detached worktree at the staging tip with nothing staged and no changes:

$ scripts/pre_commit_lint.sh; echo "exit=$?"
check: logging full output to /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-clean-tip/pre_commit_lint.log
check: nothing to check (no staged files, no working-tree changes, no branch changes vs origin/litellm_internal_staging)
check: PASS
check: full log: /Users/mateo/Development/litellm/.git/worktrees/litellm-wt-clean-tip/pre_commit_lint.log
exit=0

And a run at cfbd431 where checks are in scope (this PR's own files staged plus the scripts change exercised via make check before committing) ended the same way with the ran/skipped split and a verdict, per the new tests which also cover the FAIL path

Type

🐛 Bug Fix

Caveats (if any)

Final Attestation

• The tests check the right things, including the edge cases, and regressions in the respective real-world customer use-cases are not possible after this PR

---

Note

Low Risk
Changes are limited to the local pre-commit lint script and its unit tests; no production runtime, auth, or data paths are touched.

Overview
make check (scripts/pre_commit_lint.sh) no longer finishes with a silent exit 0 and an empty log when nothing runs. Every run now ends with a check: summary block that labels each of the four gating checks as ran or skipped (with a short reason).

When staged or scoped files match no lint gate, the script prints a NOTE listing those paths and stating the pass is a no-op, not a lint verdict. Early exits (nothing to check vs. cannot resolve merge base) also print check: PASS or check: FAIL so the outcome matches the exit code.

Tests in test_pre_commit_lint.py assert the new verdict lines and add coverage for the summary, the no-match no-op path (including a non-empty log), and failing runs.

^{Reviewed by Cursor Bugbot for commit 3fbe40c. Bugbot is set up for automated code reviews on this repo. Configure here.}

[greptile-apps]

Greptile Summary

The PR adds explicit verdicts to informational exits and appends a ran/skipped summary and final status to normal lint runs.

• Reports which Python, e2e, dashboard, and API-type synchronization checks ran or were skipped.
• Explains no-op scopes and records the summary in the existing lint log.
• Adds tests for successful, failing, empty-scope, and unmatched-file runs.

Confidence Score: 5/5

The PR appears safe to merge.

No blocking failure remains.

Important Files Changed

Filename	Overview
scripts/pre_commit_lint.sh	Adds per-check summary output, no-op diagnostics, and PASS/FAIL verdicts while preserving the existing exit status.
tests/test_litellm/test_pre_commit_lint.py	Extends the sandboxed script tests to verify summaries, no-op logging, and success/failure verdict output.

_{Reviews (2): Last reviewed commit: "fix(scripts): print the make check verdi..." | Re-trigger Greptile}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[mateo-berri]

@greptileai

[mateo-berri]

bugbot run

[cursor]

✅ Bugbot reviewed your changes and found no new issues!

Comment @cursor review or bugbot run to trigger another review on this PR

^{Reviewed by Cursor Bugbot for commit 3fbe40c. Configure here.}