Build with your coding agent
One prompt. A production-ready multi-agent backend — scaffolded, wired, and running under docker compose.
One prompt. A production-ready multi-agent backend — running under docker compose with a curl to test it.
Works in Claude Code, Codex, Gemini CLI, OpenCode, Aider, Windsurf, and Cursor.
1. Install
curl -fsSL https://agentfield.ai/install.sh | bashDrops the skill into every coding agent on your machine. In Claude Code it also ships a /agentfield slash command so you can trigger it explicitly.
2. Fire the skill
In Claude Code:
/agentfield a claims processor with risk scoring and human approvalOr just describe the system in plain English — the skill auto-matches, no slash command needed:
Build a claims-processor agent with risk scoring, pattern detection,
and human approval for low-confidence decisions.Build a research agent that spawns parallel investigators and recurses
into deeper sub-questions until the answer has citation-grade provenance.Build a compliance reviewer for support transcripts — extract claims,
check each against policy, flag violations, emit a signed audit trail.Build a security-audit pipeline that traces every vulnerability from
source to sink, then adversarially verifies each finding before reporting.3. Stack is live — hit it with curl
The skill scaffolds the agents, writes docker-compose.yml, brings the stack up, runs a smoke test, and hands you a curl command you can paste into a terminal right away.
curl -X POST http://localhost:8080/api/v1/execute/claims-processor.evaluate_claim \
-H "Content-Type: application/json" \
-d '{"input":{"id":"CL-42","description":"Fender bender, $2.4k"}}'Prefer to hand-write it? → Quickstart