runeward

The open-source governance harness for AI agents.

Put enforceable policy, human approvals, isolated execution, budgets, and signed evidence around any AI agent. Runeward works with an existing agent or multi-agent framework rather than requiring a new model or orchestration stack.

Prove it in one command

Prerequisites: a running Docker/Podman engine and the runeward binary.

runeward quickstart

The command creates .runeward/quickstart.toml, checks the policy, runtime, image, and state path, runs an allowed command, proves a destructive command is denied before execution, and verifies the signed audit trail. It never overwrites an existing policy unless --force is passed.

doctor and dashboard readiness also resolve required secret sources. A Charter that references an unset env:// value is not presented as launch-ready.

runeward doctor quickstart                     # explain setup problems safely
runeward --config-dir .runeward serve          # dashboard + governed REST API
runeward evidence export quickstart -o run.json
runeward evidence verify run.json              # independent policy/audit verification

What Runeward adds

Every governed action follows one path:

agent request → policy → human approval when required → limits → sandbox → signed audit event

Naming

Documentation and UI use familiar terms first. Existing API paths and file fields retain the original themed names for compatibility.

See the full naming and writing convention.

Install

Choose the package that matches how you use Runeward:

For normal local use, install the CLI with Homebrew. For an agent integration, install the SDK for its language as well. The pip and npm packages connect to a running Runeward API; they do not replace the CLI/runtime.

Homebrew — CLI

Local sandboxes require a running Docker, OrbStack, or Podman engine.

brew install Runewardd/tap/runeward
runeward version
runeward quickstart

pip — Python SDK

Requires Python 3.9 or newer. The base client has no third-party runtime dependencies.

python -m pip install runeward
python -c "import runeward; print(runeward.__version__)"

npm — TypeScript SDK

Requires Node.js 18 or newer.

npm install @runeward/sdk
npm ls @runeward/sdk

See Adapters for LangChain, CrewAI, LlamaIndex, OpenAI Agents, Strands, Vercel AI SDK, and LangChain.js installation options.

Other CLI installation options

The signed macOS/Linux installer requires cosign so it can fail closed while verifying the checksum manifest. Windows binaries are available from Releases.

curl -fsSL https://raw.githubusercontent.com/Runewardd/runeward/main/install.sh | sh

To build the current main branch, use Go 1.26.6:

git clone https://github.com/Runewardd/runeward
cd runeward
go build -o bin/runeward ./cmd/runeward
./bin/runeward version

Use it with an agent

Expose governed tools to an MCP-capable IDE or agent:

{
  "mcpServers": {
    "runeward": {
      "command": "runeward",
      "args": ["mcp", "--config-dir", ".runeward"]
    }
  }
}

Or place an agent CLI inside a sandbox and run one or many governed workers:

runeward cohort --agent claude --model sonnet build "Build a tested API"

Adapters are included for LangChain, CrewAI, LlamaIndex, OpenAI Agents, Strands, Vercel AI SDK, and LangChain.js. See Adapters and agent groups.

The dashboard also has a read-only Live chat TTY for each Citadel. Agent harnesses publish user, assistant, tool, and system turns with runeward_publish_conversation (or the REST, Python, and TypeScript equivalents), and authorized teammates can follow the redacted conversation without terminal input access. Runeward cannot infer private UI chat text that the agent client does not publish; wire the publish call into the harness turn callback. The publisher must connect to the same runeward serve control plane as the dashboard (through /mcp, REST, or an SDK).

Harness agents and subagents

Runeward is the enforcement boundary around an agent, not the component that decides how the agent reasons. Route the tool calls of a parent agent and each delegated subagent through Runeward to give them explicit policy, approval, isolation, budget, and evidence boundaries.

Existing concepts keep their meaning: a Cohort is a group of peer workers sharing a task board; it is not being renamed to “subagents.” The orchestrator still decides when to delegate, while Runeward records the parent/run/provider lineage and prevents a child Citadel from widening its parent's tenant or Charter. Every participating agent can receive its own Citadel and Chronicle. See Agent harnessing.

Policy workflow

Policies support built-in glob rules, CEL, OPA/Rego, and signed OCI bundles. Test them in CI, start from a reviewed scaffold, or derive exact proposals from verified production evidence:

runeward policy scaffold package-approval
runeward policy test quickstart --case 'tool=shell,action=rm -rf /,expect=deny'
runeward policy learn run.json > proposed-policy.toml

policy learn never edits a policy automatically. It verifies the evidence first, skips redacted actions, produces exact matches, and requires a human to review and broaden them.

Security posture

• The server binds to loopback by default and requires authentication before a non-loopback bind.
• Non-loopback HTTP also requires TLS unless --allow-insecure-http explicitly acknowledges that a trusted reverse proxy terminates TLS.
• Multi-principal RBAC scopes sandboxes, agent groups, recovery snapshots, and dashboard views to their tenant while attributing every operation to its actor. Static tokens and OIDC JWTs use the same authorization model, and embedded HTTP MCP shares the REST ownership checks.
• Browser automation is experimental and disabled by default. Enable it only in a trusted deployment with RUNEWARD_ENABLE_EXPERIMENTAL_BROWSER=1 after reviewing the security model. Browser-capable Charters declare capabilities = ["browser"]; the dashboard then exposes governed rendered-text and screenshot actions and their policy/egress results.
• An optional browser IDE (code-server in-cell + ticketed reverse proxy) is similarly experimental: RUNEWARD_ENABLE_EXPERIMENTAL_IDE=1, Charter [ide], Dockerfile.ide targets ide / ide-agents, examples ide-demo / ide-claude / ide-codex / ide-cursor. Limits: not per-keystroke policy; no Cursor/Claude Desktop/Codex GUIs in-cell; no first-class GitHub Copilot on code-server. See Browser IDE and the security model.
• Per-action policy applies to tool calls routed through the control plane (REST, MCP, dashboard file/shell/code actions, and SDKs). An interactive terminal or a process already running inside a sandbox is a direct sandbox session: it receives isolation/network/resource controls and terminal recording, but its individual commands are not intercepted for approval. Use governed tool calls when command-level policy and signed verdicts are required.
• Report vulnerabilities privately using SECURITY.md. Runeward remains pre-1.0; residual limitations are tracked in ROADMAP.md.

Documentation

• Quickstart
• Policies / Charters
• REST API
• Browser IDE (experimental code-server proxy)
• Security model
• End-to-end testing
• Published site: runewardd.github.io/runeward

Contributions are welcome; see CONTRIBUTING.md. Licensed under Apache 2.0.