From ham-autocode
Prefer gstack /health for general code quality checks. This skill adds ham-autocode specific checks as supplement: drift detection, uncommitted analysis, ESM/CJS compatibility. Use when: "health check", "project health", "is the project healthy", "compilation status", "run all checks", or at the start of any new session.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ham-autocode:health-checkThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **CLI alias used below:** `ham-cli` = `HAM_PROJECT_DIR="$PWD" node "${CLAUDE_PLUGIN_ROOT:-$PWD}/dist/index.js"`
CLI alias used below:
ham-cli=HAM_PROJECT_DIR="$PWD" node "${CLAUDE_PLUGIN_ROOT:-$PWD}/dist/index.js"
Automated health assessment for the target project.
ham-cli health check
This runs 5 checks with weighted scoring:
| Check | Weight | What it does |
|---|---|---|
| Git Status | 15% | Uncommitted changes, untracked files |
| TypeScript | 25% | All tsconfig*.json compilation (multi-config support) |
| Tests | 25% | Auto-detect and run test suite |
| Dependencies | 15% | npm audit for vulnerabilities |
| Lint | 20% | ESLint or Biome if configured |
Interpret the composite score:
| Grade | Score | Action |
|---|---|---|
| A | 90-100 | Ready to ship |
| B | 75-89 | Minor issues, fix before shipping |
| C | 60-74 | Significant issues, fix required |
| D | 40-59 | Major problems, stop and fix |
| F | 0-39 | Critical state, immediate attention |
ham-cli health esm-cjs
For Electron or dual-module projects.
ham-cli health drift
Finds TODO docs that say "pending" but code shows "fixed".
ham-cli health uncommitted
Generates change summaries, risk assessment, commit split suggestions.
ham-cli health quick
health check at the start of every new sessionhealth quick before any commithealth esm-cjsnpx claudepluginhub hammercui/ham-autocodeProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.