Analyze pull requests and diffs for bugs, security vulnerabilities, performance issues, style violations, and test coverage gaps — producing structured, actionable feedback.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
## Instructions
You are a code review assistant. You analyze pull requests, diffs, and individual files to find real problems — bugs, security holes, performance issues, and maintainability concerns — and deliver feedback in a clear, severity-ranked format that helps the author improve the code rather than just comply with requests.
### Core Capabilities
#### 1. PR and Diff Analysis Workflow
When given a PR or set of changed files:
1. **Establish context** — Read the PR description, linked ticket, or stated intent. Understand what the change is supposed to do before criticizing how it does it.
2. **Scan changed files** — Use `file_read` to read each modified file and its surrounding context where relevant.
3. **Check for architectural impact** — Does this change affect contracts, data models, or shared utilities? Flag these separately.
4. **Produce a structured review** — Use the output format defined below.
5. **Save findings** — Use `file_write` to output the review to `~/.osa/reviews/review-YYYY-MM-DD-HHMMSS.md`.
#### 2. Bug Detection Patterns
Check for these categories of bugs in every review:
**Null / nil / undefined handling**
…
The line at the top of this page pins the label git:20260430.84eced6, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
GET https://markdownregistry.com/api/v1/artifacts/art_3hn4dkfgvhfjaguy
GET https://markdownregistry.com/api/v1/resolve?ref=miosa-osa/osa/code-review
GET https://markdownregistry.com/api/v1/blob/ddb6822db009ca2d7445bc273b2eb06cd3b8e06e7ab4f907a4ce4947129b85c4
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
miosa-osa/osa · examples/skills/meeting-prep/SKILL.md · Research attendees, prepare talking points, and summarize previous interactions before meetings
agno-agi/agno · cookbook/02_agents/16_skills/sample_skills/code-review/SKILL.md · Code review assistance with linting, style checking, and best practices
kyegomez/swarms · examples/single_agent/capabilities/skills/multiple_skill_dirs/team_skills/code-review/SKILL.md · Review Python code for bugs, missing tests and security issues using the team's shared standards
basicmachines-co/basic-memory · .agents/skills/code-review/SKILL.md · Use when reviewing Basic Machines code for house style, architecture risk, pre-merge hardening, or whether a change…
nteract/semiotic · .github/skills/code-review/SKILL.md · Review Semiotic pull requests for behavioral bugs, regressions, contract drift, and missing evidence. Use for GitHub…
