Automated summary
Performs an exhaustive code review using OpenAI Codex, gathering context from GitHub, Linear, and Sentry issues.
Strengths
- In-depth code analysis with context from issues and project plans
- Supports uncommitted changes, specific branches, and commits
- Integrates with GitHub, Linear, and Sentry for rich context
Limitations
- Requires the codex CLI to be installed and authenticated
- Defaults to uncommitted changes only; may prompt if none exist
- Needs environment variables for Linear and Sentry integration
Use this skill before creating a pull request or after completing feature work to get a detailed review.
Avoid using it for trivial changes or when you need a quick live review without setup.
Security analysis
SafeThe skill runs a local Python script with user-provided arguments; no destructive commands, no external downloads, no obfuscation. It uses Bash but only to execute a predefined review tool, which is a legitimate use case. No risk of data exfiltration or system damage from the skill itself.
No concerns found
How this analysis is madeExamples
Review my uncommitted changesReview my changes against main branchReview my changes referencing GitHub issue #123 and Linear PROJ-456name: review description: | Exhaustive code review using OpenAI Codex. Use when user says "review", "/review", "review my changes", or wants AI-powered code analysis. Gathers context from GitHub/Linear/Sentry issues. allowed-tools: Bash, Read, Glob, Grep
Review
Exhaustive AI-powered code review via OpenAI Codex.
When to Use
- User says "review", "/review", "review my changes"
- Before creating PR
- After completing feature work
Workflow
1. Gather Issue References
Scan conversation for:
- GitHub: #123, org/repo#123, GitHub URLs
- Linear: PROJ-123, Linear URLs
- Sentry: sentry:ID, Sentry URLs
2. Read Plan File (if exists)
Check for active plan in conversation or ~/.claude/plans/
3. Read Referenced Files
If user mentions @file.ts or specific files, include content.
4. Run Review
# Default: reviews uncommitted changes if any exist
./scripts/review.py codex \
--issues "#123,PROJ-456" \
--plan ./plan.md \
--files "src/api.ts"
# If no uncommitted changes, prompts for branch/commit to compare
Command
./scripts/review.py codex [OPTIONS]
Options
| Option | Description |
|--------|-------------|
| --base, -b BRANCH | Compare against branch |
| --uncommitted, -u | Review staged/unstaged/untracked changes (default if changes exist) |
| --commit, -c SHA | Review specific commit |
| --issues, -i REFS | Issue refs: #123, PROJ-456, sentry:ID, or URLs |
| --plan, -p PATH | Plan file for context |
| --files, -f PATHS | Additional files (comma-separated) |
| --title, -t TEXT | Commit/PR title for summary |
| --model, -m MODEL | Codex model (default: gpt-5.1-codex-max) |
Context Command
Gather context without running review:
./scripts/review.py context [OPTIONS]
| Option | Description |
|--------|-------------|
| --issues, -i REFS | Issue refs (comma-separated) |
| --plan, -p PATH | Plan file path |
| --files, -f PATHS | Additional files (comma-separated) |
| --output, -o FORMAT | json or markdown (default: markdown) |
Requirements
codexCLI installed and authenticatedLINEAR_API_KEY- (optional) Linear API key for Linear issuesSENTRY_AUTH_TOKEN- (optional) Sentry auth tokenSENTRY_ORG- (optional) Sentry org slug
Output
The script displays gathered context to stderr, then launches codex review interactively. Codex outputs its review directly to the terminal.
Examples
# Default: review uncommitted changes (or prompt if none)
./scripts/review.py codex
# Explicitly review uncommitted changes
./scripts/review.py codex --uncommitted
# Review changes against specific branch
./scripts/review.py codex --base main
# Review with GitHub issue context
./scripts/review.py codex --issues "#123"
# Review with plan file
./scripts/review.py codex --plan ./plan.md
# Gather context as JSON (without running review)
./scripts/review.py context --issues "#123" --output json
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/exhaustive-ai-powered-code-review-mm78nn1f/download' --output '.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f/SKILL.md.part' && mv '.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f/SKILL.md.part' '.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
Tools declared by the author (check availability in your agent): Bash, Read, Glob, Grep
Original excerpt: Requirements
codexCLI installed and authenticatedLINEAR_API_KEY- (optional) Linear API key for Linear issuesSENTRY_AUTH_TOKEN- (optional) Sentry auth tokenSENTRY_ORG- (optional) Sentry org slug
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f/SKILL.md' && test -s '.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/review' && cp -R '.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f' '.claude/skills/review'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f'openskills read 'exhaustive-ai-powered-code-review-mm78nn1f'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/exhaustive-ai-powered-code-review-mm78nn1f/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
ClickHouse Performance Comparison
Testing
Evaluate ClickHouse performance test results, check PR performance changes, compare with history, and assess whether a result is actionable.
Analyze CI Performance Reports
Testing
Analyze CI performance comparison reports for a ClickHouse PR. Cross-references master history to distinguish real regressions from flaky tests.
Santa Method
Testing
Multi-agent adversarial verification with two independent reviewers. Output is checked twice and iteratively fixed until both pass before shipping.