agent_listLABSCO SUMMARY
A production platform, not a notebook — and the compact endpoint is what keeps it usable from Claude.
The governance layer is the reason to run it: assistant tool calls, integration writes and git pushes go through a risk policy that auto-runs, asks, or rejects by tier, with an audit trail. The tool count is large, so the compact /mcp endpoint folds the tools into roughly 33 meta-tools for clients like Claude.ai that cap tool counts, while /mcp/full exposes them individually.
What it is
FleetQ — a self-hosted platform for building, running and monitoring multi-agent systems, with a visual DAG builder, human-in-the-loop approvals and an MCP server that exposes the platform to any LLM.
What you get
- Visual DAG workflows with 8 node types — agent, conditional, human-task, switch, dynamic-fork, do-while, compensation and sub-workflow
- Multi-agent crews with coordinator, worker and reviewer roles and weighted QA scoring
- Real-World Action governance — tool calls, integration writes and git pushes route through a per-tier risk policy of auto, ask or reject
- A credit ledger with pessimistic locking and auto-pause on overspend
- The README puts the full server at 675+ MCP tools across 45 domains, of which the documentation names 125; the compact endpoint at /mcp folds them into about 33 meta-tools for clients with tool-count limits
- Structured MCP errors with gRPC-canonical codes and retry hints, and an optional deadline_ms per tool call
- tool_ssh_fingerprints and tool_bash_policy manage SSH trust and bash risk policy from MCP
Requirements
A self-hosted deployment — a Laravel 13 monolith, run via Docker Compose. The MCP server is reachable over HTTP/SSE with a Sanctum bearer token (or OAuth2 in cloud), or over stdio with php artisan mcp:start agent-fleet for a local CLI agent. Get a token from the API or the Team Settings UI. AGPLv3. A managed FleetQ Cloud tier exists if you would rather not self-host.
Setup effort
Build from source — git clone https://github.com/escapeboy/agent-fleet-o.git
USE CASES
01
Driving the platform from Claude or Cursor
Connect over the compact endpoint so a tool-limited client sees about 33 meta-tools instead of the full set.
02
Building a multi-step agent workflow visually
Compose a DAG with conditionals, loops and human-task nodes rather than gluing Python together.
03
Keeping risky actions under human control
Route tool calls, integration writes and git pushes through the per-tier risk policy of auto, ask or reject.
04
Managing SSH trust from an agent
List or delete trusted fingerprints through the SSH-fingerprints tool.
05
Bounding how long a call can run
Pass a deadline on the MCP tool so the agent caps wall-clock time per call.
125 tools
Listed in project docs. Your client may show a different list.
agent_getagent_createagent_updateagent_toggle_statusagent_deleteagent_config_historyagent_rollbackagent_runtime_stateagent_skill_syncagent_tool_syncagent_templates_listexperiment_listexperiment_getexperiment_createexperiment_startexperiment_pauseexperiment_resumeexperiment_retryexperiment_killexperiment_valid_transitionsexperiment_retry_from_stepexperiment_stepsexperiment_costexperiment_shareworkflow_listworkflow_createworkflow_generateworkflow_validateworkflow_save_graphworkflow_estimate_costworkflow_activateworkflow_duplicateworkflow_suggestionworkflow_time_gateworkflow_execution_chainproject_listproject_createproject_updateproject_activateproject_pauseproject_resumeproject_trigger_runproject_archivecrew_listcrew_getcrew_createcrew_updatecrew_executecrew_execution_statuscrew_executions_listtool_listtool_createtool_updatetool_deletetool_activatetool_deactivatetool_discover_mcptool_import_mcptool_ssh_fingerprintstool_bash_policycredential_listcredential_createcredential_updatecredential_rotatecredential_oauth_initiatecredential_oauth_finalizeapproval_listapproval_approveapproval_rejectapproval_complete_human_taskapproval_webhook_configbudget_summarybudget_checkbudget_forecastmarketplace_browsemarketplace_publishmarketplace_installmarketplace_reviewmarketplace_categoriesmarketplace_analyticsmemory_searchmemory_list_recentmemory_statsmemory_deletememory_upload_knowledgeartifact_listartifact_getartifact_contentartifact_download_infowebhook_listwebhook_createwebhook_updatewebhook_deletetrigger_rule_listtrigger_rule_createtrigger_rule_updatetrigger_rule_deletetrigger_rule_testintegration_listintegration_connectintegration_disconnectintegration_pingintegration_executeintegration_capabilitiesassistant_conversation_listassistant_conversation_getassistant_send_messageassistant_conversation_clearbridge_statusbridge_listbridge_endpoint_listbridge_endpoint_togglebridge_disconnectbridge_renamebridge_set_routingtelegram_bot_managenotification_manageteam_getteam_updateteam_memberslocal_llmteam_byok_credential_managecustom_endpoint_manageapi_token_manageBEFORE YOU INSTALL
What to know first
Tools you get
125 tools named in the project's documentation.
INSTALLATION
Connect your client
Server setup
TERMINAL
git clone https://github.com/escapeboy/agent-fleet-o.git
cd agent-fleet-o
make installMCP Clients
Cursor
<YOUR_SERVER_URL> is the address of your own FleetQ instance — the Docker install serves it at http://localhost:8080 — and <YOUR_TOKEN> is a Sanctum token: POST /api/v1/auth/token on your instance returns one that lasts 30 days, or create a long-lived one under Team Settings → API Tokens.
CURSOR · JSON
{
"mcpServers": {
"fleetq": {
"url": "<YOUR_SERVER_URL>/mcp",
"headers": {
"Authorization": "Bearer <YOUR_TOKEN>"
}
}
}
}Claude Code
<YOUR_SERVER_URL> is the address of your own FleetQ instance — the Docker install serves it at http://localhost:8080 — and <YOUR_TOKEN> is a Sanctum token: POST /api/v1/auth/token on your instance returns one that lasts 30 days, or create a long-lived one under Team Settings → API Tokens.
CLAUDE CODE · JSON
{
"mcpServers": {
"fleetq": {
"type": "http",
"url": "<YOUR_SERVER_URL>/mcp",
"headers": {
"Authorization": "Bearer <YOUR_TOKEN>"
}
}
}
}
