Analyze code correctness, design, and efficiency with confidence-scored findings.
npx skills add https://github.com/wyattowalsh/agents --skill honest-review
Please help me install this Agent Skill. Skill: honest-review Source: https://github.com/wyattowalsh/agents/tree/main/skills/honest-review Command: npx skills add https://github.com/wyattowalsh/agents --skill honest-review
This Skill automates code review, identifying defects, design flaws, and inefficiencies with confidence scores and research-backed evidence, ensuring higher code quality and faster development cycles.
Use the honest-review skill to audit the entire codebase for potential issues.
High-intent search queries and answers about installing and using this skill.
Automating code review in CI/CD pipelines involves analyzing correctness, design, and efficiency using static analysis, then outputting findings as SARIF reports and conventional comments to flag security vulnerabilities and bottlenecks before production.
Evidence-based static analysis validates code findings by cross-referencing non-trivial issues against external research using tools like WebSearch and Context7, assigning a 0.0-1.0 confidence score to distinguish confirmed defects from unconfirmed potential flaws.
Yes, you can perform a full codebase audit to identify design anti-patterns by applying multiple creative lenses like Adversary and Inversion, ensuring comprehensive detection of structural flaws and inefficiencies across the entire repository.
Generating automated refactoring fixes from a security audit involves analyzing the codebase for vulnerabilities, offering to generate fixes based on confidence-scored findings, and applying them with previews and verification steps.
SARIF output integrates seamlessly with conventional comments for pull request analysis, translating confidence-scored static analysis findings into structured, readable feedback directly within the code review workflow.
Confidence-scored code review findings range from 0.0 to 1.0, meaning lower scores indicate unconfirmed issues requiring manual verification, while higher scores represent validated defects backed by external research evidence.
NourcineAbAutomate GitHub Actions workflow creation and optimization for CI/CD management.
semicolon-devteamTrack test iterations and outcomes for GitHub issues via issue comments.
inkeepValidate PR review findings with web search and confidence calibration.
eliottbusinessAutomate pre-commit code review with static analysis and security scanning.
