with one click
pr-review-agent-skills
pr-review-agent-skills contains 4 collected skills from BerriAI, with repository-level occupation coverage and site-owned skill detail pages.
Skills in this repository
Triage a GitHub pull request for BerriAI/litellm and decide whether it is ready for human review. Classifies each failing check as PR-related vs infra/pre-existing, factors in Greptile score and CircleCI presence, and emits a thumbs-up/thumbs-down verdict with a 5-item checklist. Use when summarizing the state of a litellm PR, deciding if it is merge-ready, or triaging CI failures. Triggers on "review this PR", "is this PR ready", "triage litellm PR", or any github.com/BerriAI/litellm/pull/<N> URL.
Iteratively fix a BerriAI/litellm PR until the litellm-bot reviewer returns a 5/5 READY verdict. Posts the PR to the bot (via Slack @-mention or its /chat/api endpoint), parses the merge-confidence card, fixes every blocker the card and drilldown call out, pushes, and re-asks. Use when a contributor wants their PR fully greenlit by the bot before requesting a human review. Triggers on "loop my PR through litellm-bot", "/litellm-loop", "make this PR pass review", or "drive PR <url> to READY".
Karpathy-style senior-engineer pre-merge review for a litellm PR as if it ships to an AI gateway at 10k+ RPS on a large codebase with many flows. Answers whether the change is safe to merge, what unintended consequences could appear under load, and whether scope matches the linked issue. Detects narrow bug-class misses, scope expansion beyond the ticket, dead/unreachable code paths, hot-path perf risk, and sink-not-source fixes. Use after CI triage says READY. Triggers on "production merge review", "safe to merge at scale", "karpathy check", or a BerriAI/litellm PR URL in stdin JSON.
Review a BerriAI/litellm PR for conformance with the repo's documented and de-facto code patterns. Pulls patterns from docs/my-website/docs/ first, then from sibling files in the diff's directories. On conflict, docs win and contradicting code is flagged as tech debt rather than precedent. Use when the user asks "does this PR follow our patterns", "is this idiomatic litellm", or pastes a github.com/BerriAI/litellm/pull/<N> URL with a pattern/convention question. Do NOT use for general PR triage or CI status — that is the litellm-pr-reviewer skill.