teim-review
Performs AI-powered code reviews on OpenStack repositories.
Install
mkdir -p .claude/skills/teim-review && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/17153" && unzip -o skill.zip -d .claude/skills/teim-review && rm skill.zipInstalls to .claude/skills/teim-review
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Run a full AI-powered OpenStack code review on the current repository. Automatically detects Zuul CI context or uses local git history. Orchestrates context extraction, commit analysis, project guidelines extraction, and code review in a single pass. Writes results to .teim-review/ in the current directory.Key capabilities
- →Run AI-powered code reviews on repositories
- →Detect CI context or use local git history
- →Extract project guidelines for review
- →Generate structured review reports in HTML and JSON
- →Normalize raw structured reports and write diagnostics
How it works
The skill orchestrates context extraction, commit analysis, guideline extraction, and code review, then writes results to a specified directory, detecting CI context or using local git history.
Inputs & outputs
When to use teim-review
- →Automate code review
- →Check repository quality
- →Generate review report
About this skill
Teim Review
Follow references/workflow.md for preparation, review, validation, and deterministic report generation.
Paths and scope
skill_rootis the directory containing this installedSKILL.md. Resolve bundled scripts, references, and schemas from that directory.project_diris the repository being reviewed, normally the current working directory. Read its project instructions from there.output_dirdefaults to<project_dir>/.teim-review/. Resolve it to an absolute path before invoking scripts.- Honor the supplied review scope. If none is supplied, use uncommitted changes when present; otherwise review the latest commit. Never silently broaden scope after a Git error.
- Use supplied prepared context in CI.
execution_mode: zuulwithfindings_only: truestops after validated findings; the CI caller owns report assembly, rendering, and publication.
Review-only boundary
Invoking this skill authorizes review and the required review artifacts only. Do not run project tests, linters, pre-commit, builds, benchmarks, or CI jobs unless the user explicitly requests them. Repository instructions describing checks for code changes do not authorize running those checks during review. Read existing tests and supplied CI results as evidence; report unavailable execution evidence as a limitation.
Do not broaden the requested revision or task scope, apply fixes, or perform unrelated audits. Read surrounding code only as needed to trace in-scope behavior. The bundled context-preparation, report-generation, and artifact schema-validation tools remain part of the review workflow.
Pass this boundary to both reviewer and validator subagents. Delegation does not expand the authorized scope.
Review execution
First inspect the harness's delegation tools and applicable permissions. When delegation is supported and permitted, you must use two distinct general-purpose subagents for the reviewer and validator roles when supported and permitted. Pass each role its bundled instructions, the finding policy, schema, prepared context, and absolute resource paths. Wait for candidate generation before starting the separate validator subagent. The parent coordinates and verifies artifacts; it must not replace either role with its own reasoning merely because the review is small or delegation takes longer. Use the native delegation tool, without requiring named agent files.
When subagents are unavailable or delegation is disallowed, execute both roles sequentially in the main thread. Write candidate findings before validating them and apply the same policy and schemas. A main-thread validation pass is not independent validation; state which execution mode was used in the final summary. Record the specific missing capability or binding permission that requires fallback. Do not infer a prohibition from generic advice against unrequested delegation: this skill explicitly requests these two subagents. Do not require an extension or stop solely because subagents are absent. If a caller requires delegated execution, report an execution failure instead of silently falling back. A failed subagent is a failed step, not evidence that the harness lacks delegation.
Use the harness's existing model settings throughout. This skill does not select models, route work to cheaper models, or require named native agents.
The finding policy owns review criteria. Read the quick rules, relevant sections of the comprehensive guide, and applicable knowledge overlays as coding references.
Limitations
- →The skill expects specific file paths for style guides and knowledge roots
- →Deterministic validation requires the normalizer to be available
- →The skill's output directory is fixed at .teim-review/
How it compares
This skill automates a complete AI-powered code review, producing structured reports and diagnostics, which differs from manual code review processes.
Compared to similar skills
teim-review side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| teim-review (this skill) | 0 | 4mo | Review | Intermediate |
| python-testing-patterns | 77 | 6mo | Review | Intermediate |
| dependency-upgrade | 26 | 8mo | Review | Intermediate |
| test-cases | 57 | 10mo | No flags | Beginner |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
python-testing-patterns
wshobson
Implement comprehensive testing strategies with pytest, fixtures, mocking, and test-driven development. Use when writing Python tests, setting up test suites, or implementing testing best practices.
dependency-upgrade
wshobson
Manage major dependency version upgrades with compatibility analysis, staged rollout, and comprehensive testing. Use when upgrading framework versions, updating major dependencies, or managing breaking changes in libraries.
test-cases
cexll
This skill should be used when generating comprehensive test cases from PRD documents or user requirements. Triggers when users request test case generation, QA planning, test scenario creation, or need structured test documentation. Produces detailed test cases covering functional, edge case, error handling, and state transition scenarios.
reviewing-code
CaptainCrouton89
Systematically evaluate code changes for security, correctness, performance, and spec alignment. Use when reviewing PRs, assessing code quality, or verifying implementation against requirements.
wcag-audit-patterns
wshobson
Conduct WCAG 2.2 accessibility audits with automated testing, manual verification, and remediation guidance. Use when auditing websites for accessibility, fixing WCAG violations, or implementing accessible design patterns.
code-coverage-with-gcov
gadievron
Add gcov code coverage instrumentation to C/C++ projects