security-code-review

Reviews code for bugs and vulnerabilities and persists findings into structured reports.

Updated Jul 13, 2026
One-click install
npx skills add https://github.com/torikhayes/aeai-dotnet-brownfield --skill security-code-review-torikhayes
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: security-code-review
Source: https://github.com/torikhayes/aeai-dotnet-brownfield/tree/main/.claude/skills/security-code-review
Command: npx skills add https://github.com/torikhayes/aeai-dotnet-brownfield --skill security-code-review-torikhayes

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manual security reviews are inconsistent and their results get lost. This Skill runs a repeatable static code review for concrete bugs, vulnerabilities, and likely-issue patterns, then records every finding in a persistent store so results are auditable and comparable across runs. ## Core Features & Use Cases - Scoped or full-codebase review: Review the entire repository, or limit the scan to a specific PR number, branch, or commit range using git and GitHub CLI diff resolution. - Structured findings persistence: Each finding is stored with title, description, severity (critical/high/medium/low), and exact file path plus line number via a findings-store script. - Report generation: Produces an always-current cross-capability report plus a dated, git-tracked snapshot of the code-review findings. - Use Case: Before merging a pull request, run the review scoped to that PR to catch injection flaws, missing authorization checks, or secrets in code, and archive the dated snapshot as evidence for the change. ## Quick Start Ask the assistant to run a security code review on pull request 42 and show the resulting findings report.

Frequently Asked Questions about security-code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a security code review on a pull request?▼

Invoke the skill with the PR number as the argument, for example /security-code-review 42. It resolves the changed files with gh pr diff, reviews only those files, and records any findings with severity and line numbers.

How do I scan an entire codebase for vulnerabilities?▼

Run the skill with no arguments, which sets the scope to the full codebase. The review covers the repository under src/ and tests/ and marks the run as a full scan when finalizing.

What kinds of issues does a static security code review detect?▼

It detects concrete bugs such as logic errors and race conditions, vulnerabilities like injection, unsafe deserialization, missing authentication or authorization checks, and secrets in code, plus likely-issue patterns such as missing input validation.

Can I review only the changes on a branch or commit range?▼

Yes, pass a branch name or commit range such as main..HEAD as the argument. The skill uses git diff to resolve the changed file list and reviews exactly those files rather than the whole repository.

What happens when a security review finds no issues?▼

A zero-findings result is treated as a valid outcome, not a failure. The run is still finalized and the report explicitly states that no findings were recorded for that capability.

What are the limitations of PR-scoped security reviews?▼

A PR-scoped run only examines the changed files, so it cannot auto-resolve findings in files it never inspected. Only a full-codebase run is finalized with the full flag that allows resolving older findings.