000-index — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited 000-index (Rules) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Reusable AI coding-assistant skills, prompts, and instructions.
Pick one:
This is the simplest path. Use it if you only want the SKILL.md workflows in Claude Code:
npx skills@latest add rahul-kulkarni105/skillsThe installer reads .claude-plugin/plugin.json and installs the published skills from skills/.
Use this if you want everything in Option 1, plus the extra multi-tool setup:
npx @rahulkulkarniskills/ai-skills init --ref <tag-or-commit>For local development against this checkout:
npm --prefix cli install
npm --prefix cli run build
node cli/bin/ai-skills.js init --manifest ./manifest.jsonVerify installed files:
npx @rahulkulkarniskills/ai-skills verifyFor debugging:
ai-skills --verbose init --manifest ./manifest.json
AI_SKILLS_LOG=debug ai-skills verifyFor telemetry preferences:
ai-skills telemetry status
ai-skills telemetry enable
ai-skills telemetry disableTelemetry consent is stored in your user config directory, not in the project. DO_NOT_TRACK=1 and AI_SKILLS_TELEMETRY=0 are respected.
Anonymous telemetry is opt-in. Interactive init asks once; --yes, CI, and non-TTY runs stay disabled unless AI_SKILLS_TELEMETRY=1 is set. Telemetry is sent only when a PostHog project key is configured.
Telemetry never sends secrets, absolute paths, file contents, registry URLs, Git remotes, raw command arguments, user names, host names, raw error messages, or stack traces. See docs/telemetry.md for the full schema and controls.
grill-me, convince-me, weak-spots, steelman, pre-mortem.
— shared instructions for different agents.
.cursor/rules/000-index.mdc — native Copilot and Cursor entry points.
— reusable guidance, prompt templates, and stack-specific conventions.
Some files stay at the project root on purpose because tools look for them there:
AGENTS.md, CLAUDE.md, and GEMINI.md.README.md, LICENSE, and manifest.json..claude-plugin/, .claude/, .cursor/, .github/, and .vscode/.skills/, instructions/, prompts/, and stacks/, because themodel-facing docs and installer manifests point at those paths.
VS Code is the primary target environment for the config folders listed above. The.github/folder is read by the GitHub Copilot extension (GitHub/Microsoft);.claude/by the Claude Code extension (Anthropic);.cursor/by Cursor (Anysphere — a VS Code fork, not a VS Code extension).AGENTS.md,CLAUDE.md, andGEMINI.mdare additionally read by their respective CLI tools outside VS Code.
See docs/sync-strategies.md for the tradeoffs.
This repo holds shareable, objective content (conventions, stack guidance, adversarial skills, token discipline). Personal, subjective preferences (e.g. "I like terse responses") belong in your global config: ~/.claude/CLAUDE.md, ~/.codex/AGENTS.md, ~/.gemini/GEMINI.md, Cursor user rules, Copilot personal instructions.
Project config in a consuming repo always wins over the defaults here. See docs/working-with-ai.md.
Dual-licensed by component type:
docs — everything outside cli/): CC BY 4.0. Creative Commons explicitly recommends against CC licenses for software, so the CLI is licensed separately.
cli/ — the TypeScript installer published as@rahulkulkarniskills/ai-skills): MIT.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.