mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-15 23:31:29 +00:00
75b4478242
The workflow now runs on pull_request_target so the base branch's copy of the script and config scores every PR, forks included, and a PR can no longer edit the gate that scores it. The PR head is only ever read as a diff. Forks are passed --from-fork so the author factor is live The diff parser reads git renames, so a pure move counts zero lines and both paths take part in the paths and modules factors. The size factor counts only files outside the docs, tests, and model map tiers, and .only( joins the silenced-test markers Config: every pyproject.toml is always-human, the MCP auth, credential, OAuth, and discovery files are always-human, and the lint budget files are low. The last 400 staging merges now score 217 high, 139 medium, 44 low |
||
|---|---|---|
| .. | ||
| _agent_shin_actions.py | ||
| agent_shin_shared.py | ||
| assert_ci_coverage.py | ||
| assert_workflow_dir_hygiene.py | ||
| auto_update_price_and_context_window_file.py | ||
| close_low_quality_prs.py | ||
| detect_changes.sh | ||
| e2e_egress_sentinel.py | ||
| e2e_fetch_fixture_bundle.sh | ||
| e2e_pack_fixture_bundle.sh | ||
| risk_tier.py | ||
| run_llm_translation_tests.py | ||
| scan_keywords.py | ||
| select_ui_test_scope.sh | ||
| smoke_test_native_wheel.py | ||
| triage-requirements.txt | ||
| triage_with_llm.py | ||
| uv_sync_with_retries.sh | ||
| verify_linux_native_wheel.py | ||