from my-cc-harness122
Assess and score the overall health of a Claude Code harness across architecture, skills coverage, hooks, rules, MCP servers, eval pipelines, and team setup.
This skill performs a structured audit of a Claude Code harness, scoring eight distinct dimensions (CLAUDE.md quality, skills coverage, agents architecture, hooks automation, rules structure, MCP server capabilities, evaluation pipelines, and multi-agent/team setup). It produces a numerical score (0–24) and a graded rating (S/A/B/C/D) with actionable improvement recommendations.
Use this skill when onboarding or reviewing an agent harness, before production deployments, or during regular maintenance to identify gaps in architecture, automation, or testing. It is triggered by requests related to harness audits, configuration checks, or health diagnostics.
Designed for Claude Code harnesses and similar agent frameworks that expose CLAUDE.md, skills, and configuration files for inspection. It assumes the auditing agent can read local harness files and compute scores programmatically.
Korean-language skill for auditing Claude Code harness health across 8 dimensions (CLAUDE.md, Skills, Agents, Hooks, Rules, MCP, Eval, Team). No scripts included — purely instructional SKILL.md with scoring rubric. The skill reads ~/.claude/ directory to assess setup quality. No security concerns beyond reading local config. Content is entirely in Korean which limits accessibility. Instructions are clear within scope but quite thin — no scripts, no references/, no output contracts beyond a template.
Niche skill targeting Claude Code harness users. Well-structured rubric (0-3 per dimension, 8 dimensions) but the skill is just a static checklist with no automation. No scripts, no references folder. The allowed-tools frontmatter restricts to Read, Bash, Grep, Glob which is appropriate. Korean-only content significantly limits audience.
