ReadyAgents Core

ReadyAgents runs YAML agent workflows locally. Every node is checkpointed, so a run pauses for human approval, resumes from where it stopped, and replays offline as a regression test. Your keys, your machine, no daemon.

Site: readyagents.dev. Repo: github.com/readyagentsdev/readyagents-core.

Tried it? Open an I-ran-this issue. We are not launching. We are listening.

This repository is the free core. You keep the provider account and the bill. Install with pip install readyagentsdev, or from this clone.

60-second start

Requires Python 3.11–3.14 on Linux, macOS, or Windows. Current version is 2.0.10. Install with pip install readyagentsdev, or from this clone. The 2 versions the core contract only; extras carry their own maturity tiers.

pip install readyagentsdev
readyagents new my-flow
readyagents run my-flow/workflow.yaml
readyagents runs list

The wheel ships the example workflows too — no clone needed:

readyagents new f --from-example calc_pipeline
readyagents run f/workflow.yaml

readyagents new --list-examples shows all of them.

Or from a clone:

git clone https://github.com/readyagentsdev/readyagents-core.git
cd readyagents-core
python -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -e .
readyagents run examples/calc_pipeline.yaml
readyagents runs list
readyagents doctor

readyagents run examples/calc_pipeline.json is the same graph.

HITL next: docs/first-ten-minutes.md.

What it does

• Define agent workflows as YAML or JSON (nodes + edges)
• Run agent, tool, condition, transform, approval, parallel, include, foreach, and decide nodes. Agent nodes may declare a tools: allowlist for a bounded tool-use loop. type: decide is experimental; see docs/decisions.md.
• Persist after every node and resume a paused or failed run from the last successful node
• Inspect past runs: readyagents runs list / show / replay / fork / diff / freeze / report (local HTML)
• Scaffold a starter: readyagents new my-flow (basic, approval, research, pipeline, review, foreach, agent-tools, gated, decide)
• Builtin tools with zero extra servers: now, calc, json_get, list_dir, read_file, write_file
• MCP client and server (readyagents mcp serve, readyagents mcp probe) with official tasks and MRTR approvals
• Policy, spend ledger, budgets and caps: per-node token/cost, --estimate / --max-spend caps, model fallback, JSON logs

Beyond the core, 40+ opt-in extras ship in the same install — teams, studio, browser, knowledge, distillation, registry, environments and more. Each is listed with its maturity tier and its limits in docs/extras.md. None of them is required, and none of them changes how the core behaves.

The agent firewall (taint, tool policy, MCP pinning — security model, policy) is defence in depth, not a solution to prompt injection.

Architecture

flowchart LR
  YAML[Workflow YAML/JSON] --> Engine
  subgraph Core["ReadyAgents Core"]
    Engine[Workflow engine]
    Tools[Builtin tools]
    LLM[BYOK LLM providers]
    MCP[MCP client / server]
    Packs[Pack loader]
  end
  Engine --> Tools
  Engine --> LLM
  Engine --> MCP
  Packs --> Engine
  Packs --> Tools
  LLM --> OpenAI[OpenAI]
  LLM --> Anthropic[Anthropic]
  LLM --> Compat[OpenAI-compatible]

CLI