Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From omni
Manages Omni CLI configuration and access: auth/login/logout/recover, providers (genie/claude-code/a2a/ag-ui/openclaw/agno/webhook), API keys, status, logs, dead letters, payloads, shell completions via bash(omni *) and jq.
npx claudepluginhub automagik-dev/omni --plugin omniHow this skill is triggered — by the user, by Claude, or both
Slash command
/omni:omni-configThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use this for runtime configuration and access management (not infrastructure bootstrap).
Manages AI/agent providers via omni CLI: list/get details, create (genie, claude-code, a2a, ag-ui, agno, openclaw, webhook schemas), update/delete, health tests, Agno resources.
Quick-reference for oh-my-claudecode (OMC) v4.10+: architecture overview, mode decision tree (autopilot, ralph, team, ultrawork), agents, skills, pipelines, and strategies.
Rewires which AI CLI handles each role in cc-multi-cli-plugin, swaps CLIs, adds/disables subagents, and diagnoses CLI quirks via env vars and config files.
Share bugs, ideas, or general feedback.
Use this for runtime configuration and access management (not infrastructure bootstrap).
omni auth login --api-key <key> --api-url http://localhost:8882 --json
omni auth status --json
omni auth logout --json
omni status --json
# Recover API key when keyValid shows "no" (requires local PM2 access)
omni auth recover --api-url http://localhost:8882 --json
# Generate a new key instead of recovering the existing one
omni auth recover --rotate --json
omni config list --json
omni config get defaultInstance --json
omni config set defaultInstance <id> --json
omni config set format json --json
omni config unset defaultInstance --json
Available schemas: agno, webhook, openclaw, claude-code, genie, a2a, ag-ui
For full provider management (list, get, create, update, delete, setup, test), see the omni-providers skill.
Connects Omni to a Claude Code team inbox (Automagik Genie orchestrator).
omni providers create \
--name "genie-prod" \
--schema genie \
--agent-name "omni-agent" \
--target-agent "team-lead" \
--team-name "omni-{chat_id}" \
--json
Flags: --agent-name (required), --target-agent (required), --team-name (template with {chat_id}, {thread_id}, {sender_id}; default: omni-{chat_id})
Runs Claude Code SDK locally as an agent backend.
omni providers create \
--name "claude-dev" \
--schema claude-code \
--project-path /home/user/myproject \
--max-turns 10 \
--permission-mode acceptEdits \
--model claude-opus-4-5 \
--system-prompt "You are a helpful assistant." \
--json
Flags: --project-path (required), --max-turns, --permission-mode (default|acceptEdits|bypassPermissions|plan), --model, --system-prompt
Agent-to-Agent protocol provider (Google A2A standard).
omni providers create \
--name "a2a-agent" \
--schema a2a \
--base-url https://a2a-server.example/api \
--api-key <key> \
--json
Agent-UI protocol provider (CopilotKit AG-UI standard).
omni providers create \
--name "agui-agent" \
--schema ag-ui \
--base-url https://agui-server.example/api \
--api-key <key> \
--json
# openclaw (WebSocket gateway)
omni providers create --name "openclaw-prod" --schema openclaw \
--base-url wss://gateway.example/ws --api-key <key> --default-agent-id <agentId> --json
# agno / webhook (HTTP API)
omni providers create --name "agno-prod" --schema agno --base-url https://api.agno.com --api-key <key> --json
omni keys create --name "agent-key" --scopes messages:read,instances:write --instances <id1,id2> --json
omni keys list --status active --limit 50 --json
omni keys get <id> --json
omni keys update <id> --rate-limit 120 --expires 2026-12-31T23:59:59Z --json
omni keys revoke <id> --reason "rotation" --json
omni keys delete <id> --json
omni settings list --json
omni settings list --category ai --json
omni settings get <key> --json
omni settings set <key> <value> --reason "ops update" --json
omni start --json
omni stop --json
omni restart --json
omni update -y --no-restart --json
omni logs
omni logs error --limit 50
omni logs --modules api,nats --follow
omni logs --process nats
Flags: [level] (debug, info, warn, error), --modules <modules>, --limit <n> (default: 100), --process [service] (default: api), --follow
omni dead-letters list --status pending --limit 50 --json
omni dead-letters get <id> --json
omni dead-letters stats --json
omni dead-letters retry <id> --json
omni dead-letters resolve <id> --note "fixed manually" --json
omni dead-letters abandon <id> --json
omni payloads list <eventId> --json
omni payloads get <eventId> webhook_raw --json
omni payloads get <eventId> agent_response --json
omni payloads delete <eventId> --reason "cleanup" --json
omni payloads config --json
omni payloads config message.received --retention 30 --store-webhook true --json
omni payloads stats --json
Payload stages: webhook_raw, agent_request, agent_response, channel_send, error
omni completions bash
omni completions zsh
omni completions fish
--json + jq for automation-safe parsing.omni providers setup is interactive; avoid it in automated agent flows.omni auth recover requires local PM2 access (run on the server hosting Omni).