Community CLI alpha
Run AgentReady from a terminal before contracts reach CI.
The published Community CLI scans local OpenAPI and MCP files, writes Markdown and JSON outputs, and exits with policy status. It does not call live APIs, MCP servers, LLMs or a hosted TimeProofs backend.
Install and version
Use the alpha tag explicitly.
Community is available for free. The current alpha must be installed or run with the explicit @alpha tag. npm currently also exposes latest: 0.1.0-alpha.0 as a documented temporary exception, but new prereleases must use alpha.
npm install @timeproofs/agentready@alpha
npx @timeproofs/agentready@alpha --help
npx @timeproofs/agentready@alpha --versionImplemented command reference
The CLI exposes scan, simulate and report commands.
These forms are verified against the current CLI help output. This page does not document unimplemented account, login, upload, Pro or hosted-scanner commands.
| Command | Purpose | Outputs |
|---|---|---|
agentready scan openapi <file> | Static OpenAPI scan with optional --out, --min-score and --fail-on. | agentready-report.md and agentready.json. |
agentready scan mcp <file> | Static MCP tools JSON scan with the same policy options. | agentready-mcp-report.md and agentready-mcp.json. |
agentready simulate <agentready.json> <scenario.json> | Static scenario simulation against a previously generated AgentReady result. | agentready-simulation.json. |
agentready report <agentready.json> | Read an existing AgentReady JSON file and print score, status, tool count and risk counters to stdout. With --json, print a JSON object containing summary and tool_count. | Stdout only; this command does not write a Markdown file. |
CI policy semantics
The recommended Community gate is --min-score 75 --fail-on critical. Exit code 0 means scan and policy passed. Exit code 1 means the scan was valid but policy failed. Exit code 2 means invalid input or CLI usage error. Exit code 3 means unexpected internal error.
npx @timeproofs/agentready@alpha scan openapi ./openapi.json --out ./agentready-output --min-score 75 --fail-on critical
npx @timeproofs/agentready@alpha scan mcp ./mcp-tools.json --out ./agentready-output --min-score 75 --fail-on criticalTimeProofs AgentReady does not guarantee that an AI agent will never fail. It identifies structural risks that may cause AI agents to misuse APIs, tools or MCP servers.
A local CLI PASS means the selected static engine, ruleset and policy did not detect a covered blocking risk in that input at scan time. It is not a runtime safety guarantee.
AgentReady at a glance
The AgentReady CLI reads local OpenAPI, MCP or AgentReady JSON files and reports scan or policy results without a hosted scanner. AgentReady is a product and candidate open standard for static pre-deployment analysis. It is not a runtime firewall, independent certification, official standards-body standard or guaranteed-safety system.
- Availability
- AgentReady Community is free and available. AgentReady Pro is planned and not purchasable.
- Language
- English is the current canonical language. No translated alternate route is published for this page.