agentsfleet is a resident engineer for support escalations. A ticket arrives, an agent wakes — reads your code, telemetry, internal docs, and live control-plane state — finds the root cause, and proposes a remediation. A human approves, then it ships the fix or drafts the customer reply. Every step is a replayable log.
- Human approval, by design — the agent investigates and proposes; a person approves before anything ships
- Replayable event logs — audit every action and decision
- Bring your own provider keys — no vendor lock-in on inference
- Runs locally or against production — same agent, same evidence
Agents are defined in Markdown playbooks with tools, triggers, and investigation steps. We build the engineer, not a wrapper — open-source runtime, hosted control plane.
npm install -g @agentsfleet/cli
agentsfleet loginDefine an agent in Markdown, connect a webhook, and get an evidenced diagnosis and a proposed fix on your next escalation. Full walkthrough at docs.agentsfleet.net/quickstart — free to try, no card, under five minutes.
| Directory | What |
|---|---|
src/ |
Zig backend — agentsfleetd control plane (HTTP, leases) + agentsfleet-runner execution daemon |
ui/packages/app/ |
Dashboard — Next.js, Clerk auth |
ui/packages/website/ |
Marketing site — agentsfleet.net |
ui/packages/design-system/ |
Shared UI components |
cli/ |
Command-line interface (CLI) — install, manage agents, tail runs |
public/openapi/ |
OpenAPI spec |
schema/ |
Postgres migrations |
Prerequisites: Zig 0.16.0 · Docker (Postgres + Redis) · Bun ≥1.3 · Clerk dev project · 1Password CLI for secrets
git clone https://github.com/agentsfleet/agentsfleet.git
cd agentsfleet
# Populate .env before running make up. See playbooks/founding/01_bootstrap/001_playbook.md for the full bootstrap.
make up # Postgres + Redis + agentsfleetd (auto-migrates DB)
cd ui/packages/app
echo "NEXT_PUBLIC_API_URL=http://localhost:3000" > .env.local
bun install && bun run devVerify:
make lint-all
make test-unit-all
make test-integration # needs make up runningagentsfleet defaults to production. Point it at your local stack with the --api flag, or persist it via the environment:
agentsfleet --api http://localhost:3000 <command>
export AGENTSFLEET_API_URL=http://localhost:3000 # or set it onceEnable git hooks: git config core.hooksPath .githooks
Coding conventions and the agent operating model live in AGENTS.md.
The playbooks/ tree bootstraps your own private agentsfleet instance — a Fly.io control plane behind a Cloudflare Tunnel, with Clerk for auth and 1Password (op) for every secret. Start at playbooks/founding/01_bootstrap/001_playbook.md.
For the system design — control plane, runner fleet, data flow, scaling — see docs/architecture/, and the tunnel-first deployment rationale in playbooks/ARCHITECTURE.md.
| Repo | What |
|---|---|
| agentsfleet/agentsfleet | Control plane + runner + CLI (this repo) |
| agentsfleet/docs | User docs (docs.agentsfleet.net) |
| agentsfleet/skills | Agent skill libraries |
| agentsfleet/posthog-zig | PostHog SDK for Zig |
MIT — Copyright (c) 2026 agentsfleet.