air
01Plugin Claude Code
Plugin marketplace listing 1 plugin: air.
Plugin Claude Code
Plugin marketplace listing 1 plugin: air.
Instructions file
Instructions for VorobiovD/air, covering claude.md, what this is, project structure, architecture and development workflow.
Plugin Claude Code
Automated PR code review for GitHub — 6 specialized agents + Codex, wiki-based pattern learning, self-review/respond/re-review modes, model tiering, session-scoped /tmp, pre-commit drift-check hook, and Managed Agent CI option.
Agent
Review code changes for quality, design, test coverage, and project conventions. For security checks, use security-auditor.
Agent
Multi-agent orchestrator for managed reviews. Delegates the in-scope specialists in parallel (4 core + an optional UI/copy reviewer), then verifier, then writes wiki — mirrors the local CLI's Claude Code orchestrator on the research-preview multi-agent path.
Agent
Review code changes through the lens of git history — blame, churn, previous PR feedback, and authorship patterns.
Agent
Verify code review findings against actual source code. Filter false positives, score confidence, and confirm real issues.
Agent
Audit code changes for security vulnerabilities, data exposure, injection risks, auth gaps, and compliance concerns.
Agent
Review changed code for reuse, quality, and efficiency. Report findings only.
Agent
Review user-facing copy and statically-detectable UX/accessibility in UI changes — flag developer jargon, AI-generated fluff, and clarity/a11y problems. Report findings only.
Command
Clean up, deduplicate, and reorganize the wiki REVIEW.md using AI. Also regenerates REVIEW-HISTORY.md from PR comment history.
Command
When --respond is passed, this flow automates the developer's side of the review cycle. It reads the existing review, classifies each finding based on local code changes, verifies fixes are correct, runs a self-check on the fix diff, detects additional changes, and posts a structured response that the reviewer's next…
Command
When --self is passed, this is a completely different flow. No PR needed. Reviews local changes before push.
Command
One agent applies all six review lenses + self-verifies in a single pass — the CLI counterpart of managed's AIRREVIEWMODE=solo, sharing the same assembled prompt (lib/soloprompt.py) and the same verdict-gating contract (lib/verdict.py). Runs on Fable via your Claude Code subscription: 3–7 min of agent time and $0 of…
Command
Automated code review with verification, pattern learning, and team knowledge — review PRs, self-check before pushing, or track fixes across iterations.
Hook
Runs before the agent uses a tool for Bash tool calls, executing pre-commit-drift.py via python3. From VorobiovD/air.