Flightdeck.

Self-hosted · Apache 2.0 · Production and coding agents

Observability and control plane for AI agent fleets.

Every LLM call, tool call and MCP call your coding agents and production agents make streams to one dashboard as it happens, and from that dashboard you set token budgets, allow or block MCP servers, and stop an agent.

Linux & macOS
curl -fsSL https://flightdeck-hq.com/install.sh | sh

Then run flightdeck and open http://localhost:4000.One binary, nothing else to install.

Every channel installs the same binary, for Linux, macOS and Windows on amd64 and arm64. Release page (opens in a new tab)

Fleet viewStreamingPaused
research-agent
support-triage
claude-code · omri
etl-summarizer
llm_call tool_use mcp_event policy_warn / degrade policy_block / kill

One row per agent, sub-agents nested under their parents, events plotted as they arrive. Hover any event to inspect it; hovering pauses the stream.

What it does

Live fleet view

Every agent on a shared timeline: LLM calls, embeddings, tool uses, MCP events, and structured errors, streamed as they happen. Click any event to inspect it.

Budgets that act

Central token budgets per flavor, enforced in-process. Warn at 80%, swap to a cheaper model at 90%, block at 100%. Defaults, all configurable, every decision a structured event.

Kill switch and directives

Stop one agent or an entire flavor from the dashboard; the signal delivers on the agent's next call. Register custom directives in Python and invoke them remotely.

MCP protection

Allowlist or blocklist the MCP servers your agents may talk to, per flavor. Misconfigured or unrecognized servers warn or block before the call leaves the process.

Sub-agent trees

Multi-agent runs render as a tree: orchestrator parent, one child session per sub-agent, linked and labeled by role. Claude Code tasks, CrewAI agents, LangGraph nodes.

Analytics and cost

Token consumption, latency, model distribution, policy volume, and estimated cost, grouped by flavor, model, framework, host, team, or role.

Architecture

Not a proxy. The sensor wraps SDK clients inside your agent's own process: init() + patch(), no other code changes, and calls go straight to the provider. Prompt capture is off by default in the sensor; the Claude Code plugin captures your own session by default and can be turned off.

Anthropic OpenAI litellm LangChain LangGraph LlamaIndex CrewAI bifrost Claude Code MCP

On the dashboard

Live fleet viewStreaming

The live fleet feed. Agents, events, run boundaries, and policy decisions on one shared clock.

Agents

Every agent with token, latency, error, and cost trends, plus a per-agent drawer and swimlane.

Events

Filter every event by agent, type, framework, model, or MCP server. Click through to the run.

Quickstart

1Install and run Flightdeck

Linux and macOS

curl -fsSL https://flightdeck-hq.com/install.sh | sh

Windows (PowerShell)

irm https://flightdeck-hq.com/install.ps1 | iex

Or the same binary another way

Homebrew

brew install flightdeckhq/tap/flightdeck

pip (PyPI)

pip install flightdeck-cli

The same binary, as the flightdeck command.

uvx (PyPI)

uvx flightdeck-cli

Runs it from PyPI without installing.

Docker

docker run -p 127.0.0.1:4000:4000 -v flightdeck:/data flightdeckhq/flightdeck

The image creates an access token on its first start and prints it; agents need it as FLIGHTDECK_TOKEN.

Then start it with flightdeck. The dashboard is at http://localhost:4000, and nothing needs configuring: agents on the same machine find it without a server address or a token. flightdeck --port 4100 runs it on another port, where flightdeck connect claude-code finds it, and flightdeck --public lets other machines reach it; whoever can reach it can then read its token, so open it only on a network you trust, or put it behind a proxy that authenticates. Every channel installs the same binary, for Linux, macOS and Windows on amd64 and arm64; the archives are on the release page, and guides/windows.md has the rest for Windows.

2Connect your agents

Claude Code

The installer offers this when it finds Claude Code; otherwise:

flightdeck connect claude-code

Sessions you start from then on appear in the fleet view within seconds. It needs a recent Claude Code: with an older one it says to run claude update. Or install the plugin from inside Claude Code:

/plugin marketplace add flightdeckhq/flightdeck
/plugin install flightdeck@flightdeck-plugins
/reload-plugins

Python agents

Install the sensor, then add these lines at the top of your entrypoint:

# pip install flightdeck-sensor
import flightdeck_sensor

flightdeck_sensor.init()
flightdeck_sensor.patch()

Nothing else in your code changes. Every call through the Anthropic or OpenAI SDK or litellm, directly or inside a framework, is reported, with prompt capture off until you turn it on.

For a Flightdeck on another machine, set FLIGHTDECK_SERVER to its address and FLIGHTDECK_TOKEN to a token minted under Settings → Access tokens: the plugin and the sensor take the same two values. For Claude Code, flightdeck connect claude-code --server <address> --token-file <file> saves both in the plugin’s options instead.

Docker Compose and a Helm chart run the same binary in production; the self-hosting guide has both. Full walkthrough in the README.