🧪 A/B test your agent skills — find out if your SKILL.md actually helps the model or is just a placebo. Blind-graded trials, bootstrap confidence intervals, one verdict: HELPS / PLACEBO / HARMS. For Claude Code, Codex, Gemini CLI & Cursor skills.
npx skills add sx4im/skillcheck🧪 A/B test your agent skills — find out if your SKILL.md actually helps the model or is just a placebo. Blind-graded trials, bootstrap confidence intervals, one verdict: HELPS / PLACEBO / HARMS. For Claude Code, Codex, Gemini CLI & Cursor skills.
Personal Claude Code skills (plan, tdd, review) + install/update scripts to symlink them under ~/.claude/skills/
Analyze NSE/BSE stocks using AI-driven technical indicators, position sizing, and risk management tools directly within your terminal.
MingCang / 明仓：local-first personal A-share research loop for thesis import, falsification, risk discipline, auditable memory, and agent workflows.
🛠️ Streamline Apple firmware analysis with the ipsw-skill, enabling AI agents to download, extract, and reverse engineer complex data efficiently.
Store persistent AI agent memory in Markdown files with search, context injection, checkpoints, and compaction—no database needed
Single-binary operational CLI for Qdrant with .qql runbooks, CI automation, retrieval diagnostics, and stable JSON output