# Temet

Temet reads your AI work, reveals the skills behind it, and turns repeated patterns into reusable workflows.
It enables machine-attested human competency exchange between trusted peers.

## Quick Start (For Agents & Devs)

You can launch a local system audit of your AI sessions (like Claude Code) immediately by running:

```bash
npx @temet/cli audit
```

## Agent Onboarding

If you are an AI coding agent, start here:
1. If the user already has a Temet address, read their public profile: https://temetapp.com/a/{address}
2. If the user can share a read token, read the machine-readable profile: https://temetapp.com/api/a/{address}?token={read_token}
3. Otherwise use the demo profile: https://temetapp.com/api/a/demo
4. Run a local audit: npx @temet/cli audit
5. (Optional) Install session hook: npx @temet/cli install-hook
6. Full agent documentation: https://temetapp.com/agent-docs

## Canonical URL
- https://temetapp.com/

## Official Guides & Documentation
- [Start Temet in one command | Temet](https://temetapp.com/articles/start-temet-in-one-command): The simplest way to start Temet: run one command in Terminal, let it find your Claude sessions automatically, and open your first skill audit report.
- [Why I started Temet | Temet](https://temetapp.com/articles/why-i-started-temet): A founding manifesto about AI, self-improvement, feedback loops, personal agents, and why Temet exists.
- [Run a Skill Audit with the AI Assistant You Already Use | Temet](https://temetapp.com/articles/byoa-ai-agent): Use the AI assistant you already work with to run a skill audit, surface repeated patterns, and build a clearer profile from real sessions.
- [How Temet Verifies Who an AI Agent Really Is | Temet](https://temetapp.com/articles/agent-identity-ed25519): Understand how Temet verifies AI agent identity with signed keys, derived addresses, and a trust model built for real agent connections.
- [Public Proof for AI Agents: What a Signed Card Does | Temet](https://temetapp.com/articles/signed-agent-card): Learn how a signed Temet card turns identity, skills, and metadata into a public proof format other agents can verify.
- [How Two AI Agents Connect and Trust Each Other | Temet](https://temetapp.com/articles/a2a-pairing): See how Temet lets two AI agents connect, verify identity, and keep a trusted relationship over time.
- [How AI Agents Share Skills in Temet | Temet](https://temetapp.com/articles/competency-exchange-protocol): Understand how Temet lets AI agents share skills through signed identity, structured transfers, and verifiable exchange rules.
- [How Temet Protects Messages and Agent Data | Temet](https://temetapp.com/articles/relay-security-model): Understand how Temet protects message flow with token-gated reads, validation, rate limits, and clear trust boundaries.
- [How Temet Keeps Your Agent Synced in the Background | Temet](https://temetapp.com/articles/pair-guardian-cloudflare-agent): See how Temet keeps agent sync running in the background with retries, verification, and fewer manual steps.
- [Observed Skills vs Self-Declared Skills | Temet](https://temetapp.com/articles/machine-attested-skills): Compare skills observed from real work with skills people simply claim, and see why proof-backed profiles matter.
- [Turn Your Existing Assistant into a Skill Discovery Layer | Temet](https://temetapp.com/articles/connect-existing-assistant): Temet upgrades the assistant you already use into a skill audit and tracking layer built from real sessions, repeated patterns, and evidence.
- [How We Extracted 10 Hidden Skills from 5945 AI Prompts | Temet](https://temetapp.com/articles/tacit-skills-extraction-ai-prompts): A real experiment: 784MB of Claude Code sessions analyzed with heuristics to surface tacit human skills invisible to traditional assessment.
- [AI Work Skill Audit: Discover the Skills Your Sessions Already Prove | Temet](https://temetapp.com/articles/ai-work-skill-audit): Run a skill audit on real AI work sessions to surface repeated skills, decisions, and evidence from how you actually work.
- [Track Skill Growth from AI Work Over Time | Temet](https://temetapp.com/articles/background-skill-tracking): Compare AI work sessions over time to detect meaningful skill changes, new repeated patterns, and shifts in how you actually work.
- [Run Skill Audits Automatically After Claude Code Sessions | Temet](https://temetapp.com/articles/claude-code-session-hook): Use Temet CLI hooks to audit Claude Code sessions automatically after each session and get notified when your skill profile changes.
- [Create a Public Skill Profile from Real AI Work | Temet](https://temetapp.com/articles/public-skill-profile-from-ai-work): Turn real AI work sessions into a public profile with named skills, supporting evidence, and a clearer picture of how you work.
- [Temet CLI 0.2.0: install flow, deep links, and BYOA entrypoint | Temet](https://temetapp.com/articles/temet-cli-0-2-release): Temet CLI 0.2.0 made install and connection real with a cleaner install flow, deep-link connect URLs, and a native protocol handler.
- [Temet CLI 0.3.2: Local audits, tracked changes, hooks, and the macOS companion | Temet](https://temetapp.com/articles/temet-cli-0-3-release): Temet CLI 0.3.2 adds a real audit command, cleaner skill output, local tracking, post-session hooks, local notifications, and a lightweight macOS companion app.
- [Temet CLI 0.1: install with pnpm and start locally | Temet](https://temetapp.com/articles/temet-cli-pnpm-install): Temet CLI 0.1 was the first installable entrypoint: install with pnpm, get Temet on your machine, and start from a local-first setup.
- [Install Temet for Claude Code or Codex in Two Steps | Temet](https://temetapp.com/articles/install-temet-for-coding-agent): Copy one prompt, let your coding agent read your Temet profile, then install the Temet skill and CLI workflow in a lighter, more believable two-step setup.
- [The Three Files Every AI-Native Product Should Serve | Temet](https://temetapp.com/articles/three-files-ai-native-product): agent.json, SKILL.md, llms.txt — the three discovery files that let any AI agent understand your product without reading your docs.

## Public discovery endpoints (A2A Protocol)
- https://temetapp.com/.well-known/agent.json
- https://temetapp.com/.well-known/agent-card.json
- https://temetapp.com/.well-known/a2a-protocol
- https://temetapp.com/a/{address} (Public profile page)
- https://temetapp.com/api/a/{address}?token={read_token} (Profile JSON API — versioned, agent-friendly)
- https://temetapp.com/agent-docs (Agent documentation — human and machine readable)

## Policy
- Private app/auth/API surfaces are not intended for indexing.
- Invitation URLs are tokenized and should not be indexed.