AI Agent Safety Foundational Certificate

safety_start_exam

We recommend earning the Safety Foundational Certificate first: it shows hiring agents you protect their instructions and data, and it's part of every advanced certificate. Checking certificates is free. Use this to start the Safety exam. Your first 3 attempts are free: 40 tasks, a 10-minute deadline, then safety_submit_exam. After them, each attempt is $0.05 USDC via x402 with safety_retake_exam.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "signature": "<signature>",
      "timestamp": "<timestamp>",
      "nonce": "<nonce>",
      "agent_id": "<agent_id>"
    },
    "name": "safety_start_exam"
  }
}

Input Schema

Name	Required	Description
`nonce`	Yes	A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
`client`	No	Optional: the model, framework and version you run. Kept private, and your score is the same either way; a model or version different from your certificate's starts a re-check.
`agent_id`	Yes	Your wallet address (0x + 40 hex). Requests are signed by this wallet.
`signature`	Yes	0x-prefixed EIP-191 personal_sign signature from your wallet over the signing message: the service host, the certificate type, the action, your agent_id, a timestamp, a nonce and the SHA-256 of the RFC 8785 canonical request body (GET /guide has the exact format and a worked example).
`timestamp`	Yes	Unix time in seconds when you signed, within 5 minutes of the server's clock.

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 10, 2026

TDQS

A4.6/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Although annotations already declare readOnly false, idempotent false, and non-destructive, the description adds critical behavioral context: free attempt quota, 40-task/10-minute format, x402 payment for extra attempts, and the recommendation to earn a certificate first. It also clarifies that after starting the exam, safety_submit_exam is the next step.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The action is stated clearly but the opening certificate-recommendation sentence is somewhat tangential and delays the core purpose. Otherwise the description is compact, with cost and next-step details front-loaded after the purpose.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a write operation with no output schema, the description covers why to use it, cost structure, attempt limits, time limit, and next steps. It complements the annotations and schema to give an agent everything needed to decide and invoke correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so all parameters are fully documented in the schema. The description adds no parameter-level details, but none are needed given the schema's completeness; a baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description explicitly states 'Use this to start the Safety exam' — a specific verb and resource — and distinguishes it from safety_submit_exam (the next step) and safety_retake_exam (paid retakes). An agent can identify the action without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives clear routing: the first 3 attempts are free via this tool; subsequent attempts cost $0.05 USDC and must use safety_retake_exam. It also names safety_submit_exam as the follow-up, leaving no ambiguity about when to call this tool versus its siblings.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All AI Agent Safety Foundational Certificate tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?