Parse review findings and apply wave-ordered fixes with atomic commits.
npx skills add https://github.com/jhd3197/CachiBot --skill review-fix-jhd3197
Please help me install this Agent Skill. Skill: review-fix Source: https://github.com/jhd3197/CachiBot/tree/main/.claude/skills/review-fix Command: npx skills add https://github.com/jhd3197/CachiBot --skill review-fix-jhd3197
Many repositories accumulate review findings that are documented but never consistently applied, leaving technical debt and broken linters behind. This Skill automates reading the newest review, synthesizing prioritized fix actions, and applying minimal, auditable changes so maintainers can close review gaps quickly and safely.
Run the review-fix skill to read the latest review file, generate a prioritized plan, and apply fixes with atomic commits.
High-intent search queries and answers about installing and using this skill.
You can automate applying code review fixes by parsing findings from .reviews/*-review.md files, generating wave-ordered plans, and executing minimal edits with linters and atomic commits. This Skill handles backend, frontend, data, and test remediation automatically.
Turning code review findings into atomic commits requires wave-based execution that groups related fixes by severity and file scope. This Skill stages changes per wave, creating one atomic commit per wave while producing execution logs.
Yes, this code review automation works with ruff and npm linters by running them during wave-based execution. It applies minimal edits to resolve findings and validates changes using these linters before staging and committing.
You can prioritize backend, frontend, and data fixes by spawning parallel planning agents that group related findings and rank them by severity. This generates a wave-ordered review-plan ensuring high-severity issues are addressed first.
Yes, you can generate an execution log for code review remediation. The Skill outputs .reviews/YYYY-MM-DD-review-execution.md files documenting decisions, commits, and skipped items alongside a corresponding review-plan file for full auditability.
This tool requires code review findings stored as markdown files matching the .reviews/*-review.md format. It parses these files to extract structured findings including severity, file, line, and reviewer metadata for automated remediation.
j0rugeParse GitHub PR AI review comments and generate per-reviewer markdown checklists.
jontsaiParse date formats and run streakify.sh to backfill GitHub streak commits.
drillanRun Ruff and Mypy pre-commit checks to enforce code quality.
kriscendobotRun Prettier and ESLint fixers and deterministic probes on staged changes.
