Skill

Teim Review

Runs an AI-powered OpenStack code review using Zuul CI or local git history, producing a structured report with HTML output.

Stars

Invocation

How this skill is triggered — by the user, by Claude, or both

Slash command

/teim-review:teim-review

User invocable

Model invocable

Inline context

Default effort

The summary Claude sees in its skill listing — used to decide when to auto-load this skill

Use the @teim-review-agent subagent to perform a complete code review.

36 lines · ~433 tokens

Stats

LanguagePython

Stars8

MaintenanceExcellent

Last CommitJun 22, 2026

Actions

Tags

Use the @teim-review-agent subagent to perform a complete code review.

The authoritative shared workflow is prompts/teim-review-core.md. This skill is the Claude-native adapter for that workflow.

Output directory: .teim-review/ (relative to the current working directory)
Project directory: current working directory
Context detection: check whether ZUUL_CHANGE is set; if so, operate in Zuul CI mode; otherwise use local git context
Changed files: generate .teim-review/changed-files.txt by invoking the bundled deterministic helper skills/teim-review/scripts/detect_changed_files.py. In Zuul mode use the helper default; in local uncommitted review pass --uncommitted-only; when a base branch is selected pass --base-ref; pass --allow-root-commit only when intentionally treating all tracked files as changed.
Generate HTML: yes — write .teim-review/review-report.html
Style guide: use ./docs/quick-rules.md and ./docs/comprehensive-guide.md from the current repo
Knowledge root: use ./docs/knowledge/ from the current repo for overlays and example-backed review context
JSON schema: schemas/review-report-schema.json (relative to the plugin root)
Semantic profiles: config/tool-profiles.json defines shared fast and deep aliases across tools