| name | setup-archon-skills |
| description | Interactive first-run wizard walking the user through 8 configuration prompts and optionally bootstrapping the project. Use on first launch or when reconfiguring archon-cli. |
Setup Archon Skills
Interactive configuration wizard. The agent walks the user through 8 prompts naturally in conversation. The agent uses Read/Write/Bash tools to inspect and update configuration — the skill only provides the instructions.
Process
0. Read existing config
Use the Read tool to inspect ~/.config/archon/config.toml if it exists. Report current values so the user knows what they're changing.
1. Authentication
Ask: "Use OAuth (Claude.ai subscription) or API key?"
- If OAuth: tell them to run
archon login
- If API key: tell them to set
ANTHROPIC_API_KEY environment variable
2. Personality
Ask: "MBTI type for the agent? (default: INTJ)"
Write to [personality] section.
3. Theme
Ask: "Theme? (default: derived from MBTI)"
Write to [tui] section if explicit choice given.
4. Permission mode
Ask: "Default permission mode? (default | acceptEdits | auto | dontAsk | bypassPermissions)"
Write to [permissions] section.
5. Embedding provider
Ask: "Embedding provider? (auto | local | openai)"
Write to [memory] section.
6. Model
Ask: "Default model? (claude-sonnet-4-6 | claude-opus-4-8 | claude-haiku-4-5)"
Ask: "Effort level? (high | medium | low)"
Write to [api] section.
7. Output paths
Ask: "Where should /to-prd write PRDs? (default: /prds/)"
Ask: "Where should task specs go? (default: /tasks/)"
Write to [paths] section, or accept defaults.
8. Project initialisation
Ask: "Initialise this project for archon-cli now? (creates .archon/, prds/, tasks/)"
If yes:
- Check whether
./scripts/archon-init.sh exists (Path B: building from source)
9. Summary
Print a final summary of all choices and where they were saved.