Models
Model access overview for Coven: OpenAI through Codex, Anthropic through Claude Code, GitHub through Copilot CLI, local models, and no daemon credential ownership.
1 min read
Coven is not a model gateway. It does not store provider credentials, route model API calls, or mint provider tokens. A familiar reaches models through a harness.
Current model paths
| Provider path | Current posture |
|---|---|
| OpenAI | Through the Codex harness. |
| Anthropic | Through the Claude Code harness. |
| GitHub | Through the Copilot CLI harness. |
| Connected providers in Coven Code | Through the bundled Coven Code harness; credentials live in the engine's own auth store. |
| xAI | Through the grok adapter recipe; auth via grok login or XAI_API_KEY in the harness's environment. |
| Local models | Local embeddings ship today in the coven-memory archival layer; local-model agents arrive through adapter manifests. See Local models. |
What Coven sees
Coven sees harness process output and exits. It does not see provider API requests unless the harness prints them. It does not need API keys to launch a session.
coven run --model <id> is a launch convenience, not provider access: Coven strips any provider/ prefix and forwards the bare id to the harness's native model flag (for example codex --model or claude --model). The harness still authenticates and calls the provider itself.
Client rule
Clients should not pass model credentials into Coven. They should launch a harness whose provider auth is already complete and let the harness call the provider.
For the exact credential boundary, see Provider boundary and Harness provider auth.