cycgraph
Build complex multi-agent workflows that loop, branch, and recover. Cyclic graphs, durable state, zero-trust security.
Why cycgraph
Section titled “Why cycgraph”Most agent frameworks model workflows as DAGs — agent A calls B calls C. That works for simple pipelines and breaks for everything else: looping back on validation feedback, supervisors that route dynamically, populations that evolve in parallel, workflows that pause for human review and resume hours later.
cycgraph solves this with a Cyclic State Graph: nodes that can loop, revisit prior nodes, and make routing decisions by reading from a shared state object. Every state transition is auditable. Workflows survive crashes. Agents can’t see what they shouldn’t.
What you get
Section titled “What you get”- Cyclic graph engine — loops, retries, conditional routing, nested subgraphs, parallel fan-out.
- Six built-in patterns — Supervisor, Swarm, Evolution, Self-Annealing, Map-Reduce, Human-in-the-Loop.
- Durable execution — every action persisted; runs survive crashes via event-sourced replay.
- Zero-trust security — per-node
read_keys/write_keys, taint tracking on all external data, MCP server allowlist. - Budget guardrails — token, cost (USD), iteration, and wall-clock limits, all enforced at the engine.
- Production observability — OpenTelemetry tracing, structured events, real-time streaming via async iterables.
- Pluggable persistence — in-memory by default; Postgres adapter for production durability.
Get started
Section titled “Get started”npm install @cycgraph/orchestratorPick the path that fits where you are:
- Quick Start — install, set an API key, run a workflow in five minutes.
- Core Concepts — graphs, nodes, agents, state. The four primitives.
- Workflow Patterns — runnable examples of each pattern.
- Architect — generate workflow graphs from natural language.
If something breaks on the first run, Troubleshooting covers the common errors and their fixes.