Run parallel security, adversarial, and spec-compliance reviews and write a JSON report.
npx skills add https://github.com/iroh4-labs/mumei --skill peruse
Please help me install this Agent Skill. Skill: peruse Source: https://github.com/iroh4-labs/mumei/tree/main/skills/peruse Command: npx skills add https://github.com/iroh4-labs/mumei --skill peruse
This Skill automates the review pipeline for plan-vehicle features, providing a streamlined process for code review and quality assurance.
Run the 'peruse' skill to initiate a review for the current plan-vehicle feature.
High-intent search queries and answers about installing and using this skill.
Automating code review for plan-vehicle features involves running parallel security, adversarial, and spec-compliance reviews. This process validates findings, aggregates a final verdict, and writes a JSON review report to ensure quality assurance.
Parallel processing in a code review pipeline runs multiple specialized reviewers, such as security, adversarial, and spec-compliance checks, simultaneously. This validates findings concurrently and aggregates a unified verdict for faster quality assurance.
You get a JSON review report by executing an automated review pipeline that validates findings and aggregates a verdict. Upon completion of the parallel reviews, the report is written directly to your specified directory.
Yes, you can run security and spec-compliance reviews at the same time. The pipeline executes these adversarial and compliance checks in parallel, validates their findings, and aggregates the results into a single verdict.
The best way to validate findings during an automated code review is using an issue validator within the pipeline. It processes parallel security and spec-compliance outputs, verifying issues before aggregating a final verdict.
lengo0951Analyze codebase changes for security vulnerabilities, compliance, and failure modes.
Logisoft-webScan code repositories for vulnerabilities in dependencies, secrets, and configurations.
baphledRun linting, formatting, security scans, and tests before committing code.
CaptainCrouton89Evaluate code changes for security, correctness, performance, and spec alignment.
