Skip to content

v1.27.0 - Local agent API + playground

Choose a tag to compare

View all tags
@FJRG2007 FJRG2007 released this 07 Jul 10:21
v1.27.0
c43bfaf

Local OpenAI-compatible API for your coding agents

enigma api serves your local coding agents (Claude Code, and Codex/OpenCode where installed) over one OpenAI-compatible HTTP API, reachable from any OpenAI client library - all of their tools, skills, MCP and sessions, with your existing login. Loopback-only.

  • Multi-agent: one server, many backends - route per request by the model field (claude-sonnet-5 | codex | opencode).
  • Accounts / profiles / packs: run requests under a specific account, a profile's mapping, or a pack's isolated context (e.g. Helio) - via --account/--profile/--pack, a per-request body field, or a persisted default settable from the dashboard.
  • Formats: OpenAI (/v1/chat/completions) and Anthropic (/v1/messages), streaming and non-streaming.
  • Auth: optional bearer key (--api-key / ENIGMA_API_KEY).
  • Dashboard: a new Playground tab to test the API (format, backend, account/profile/pack, auth, in-process or HTTP mode, curl output).

Docs: https://fjrg2007.github.io/enigma/docs/concepts/api/

Assets 6
Loading