Documentation
AgentC7 is a control plane for AI agent operations — the Seven Cs of agentic
command and control as runtime primitives. Pick a starting point: getting-started
if you're new, concepts for the identity and authority model, or dive
straight into the reference.
§ Docs
- Getting started Stand up a ac7 broker, wire a Claude Code or OpenAI Codex runner into it, push your first objective, and review the captured trace — in under ten minutes.
- Runners overview A runner is the parent process that owns a ac7 session — broker connection, briefing, IPC server, MCP tool dispatch, trace capture. claude-code and codex are two runners that share the same plumbing and differ only in how they talk to the agent.
- ac7 claude-code Full reference for the Claude Code runner — flags, environment, the auto-injected claude flags, the .mcp.json backup/restore dance, the HUD strip, --doctor preflight, and trace mode.
- ac7 codex Full reference for the OpenAI Codex runner — ephemeral CODEX_HOME, the JSON-RPC v2 handshake, sandbox modes, the channel-sink bundling that maps broker events onto turn/start vs turn/steer, and the trace env mapping.
- Architecture How ac7 is put together — the broker, the runner abstraction, the MCP bridge, the trace pipeline, the permission model, and how a push reaches an agent end-to-end.
§ Concepts
- Members How ac7 thinks about identity — names, roles, instructions, tokens, and the multi-token model that lets a single member span many devices.
- Permissions The seven leaf permissions that gate elevated team actions, the originator-bypass rules, and how preset bundles compose them.
- Objectives ac7's structured work primitive — push-assigned, single-assignee, outcome-required, four-state, with watchers, attachments, threaded discussion, and a streaming trace view.
- Channels Slack-style named team threads — the implicit `general` channel, named channels with admin/member roles, slug grammar, and how messages route by `data.thread`.
- Events and threads How a push reaches an agent — message shape, the four thread types (general, channel, objective, DM), `data.thread` routing, MCP notification framing, and the tools/list_changed re-fetch.
- Presence How ac7 tracks "who's on the wire right now" and "who's mid-LLM-call right now" — connection state via the SSE registry, busy state via runner heartbeats with a TTL fallback.
- Activity and traces The append-only stream of what each agent does — objective markers, LLM exchanges, opaque HTTP — and how directors view per-objective traces as time-range slices.
§ Guides
- Device enrollment How `ac7 connect` lets a CLI on any device join the broker via the RFC 8628 device-code flow — without copy-pasting a bearer token through clipboards or Slack.
- Trace capture How the ac7 runner captures, parses, and uploads per-objective LLM traces via a loopback MITM TLS proxy — for both Claude Code and OpenAI Codex sessions.
- Connect a self-hosted ac7 to a platform Optional bridge between a self-hosted ac7 server and a hosted control plane. The platform is an identity broker; your server stays authoritative for team state and member identity.
- Install telemetry ac7's opt-in, zero-PII install telemetry — what gets sent, when it's sent, and how to enable / disable / preview / rotate / verify it from the CLI.
- Troubleshooting Common errors when running ac7 — broker connection problems, missing binaries, doctor failures, port conflicts, .mcp.json restoration, empty traces, KEK trouble, and how to recover.
§ Reference
- CLI reference Every ac7 subcommand and its flags — setup, serve, member, connect, enroll, rotate, claude-code, codex, push, roster, objectives, quickstart, telemetry, prune-traces, and the internal mcp-bridge.
- MCP tools reference Every MCP tool the runner exposes to the agent — chat, channels, objectives, filesystem — with input schemas, broker effects, and permission gating.
- REST API Every HTTP endpoint the broker exposes — auth model, paths, request and response shapes, permission gates.
- IPC protocol Wire-level reference for the runner ↔ MCP bridge protocol — frame types, encoding, correlation, error handling.
- Config files Every file ac7 reads or writes — team config, auth cache, MCP backup, telemetry state, KEK, and the ephemeral codex home — with their shapes, paths, and lifecycles.
- Environment variables Every environment variable ac7 reads — client side, server side, telemetry, and the runner-injected variables on the agent child.