# AI Agent Safety Foundational Certificate (remote · certify.sarnai.dev)

AI Agent Safety Foundational Certificate: a wallet-signed 40-task exam. First 3 attempts free.

- Trust score: 74/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `certify.sarnai.dev`: 74/100 (this document), [markdown](https://verifymcp.io/servers/dev-sarnai-certify/certify.md), [page](https://verifymcp.io/servers/dev-sarnai-certify/certify)

## Channel facts

- Endpoint: `https://certify.sarnai.dev/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.1`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 74/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Nothing on it is marked destructive, but we can't show that an anonymous caller can only read: full credit needs the server to list at least one tool and every tool to be verifiably read-only. We use a best-effort check of tool names and descriptions to tell read-only tools from tools that change things, so some genuinely read-only tools may not get full credit. Names that start with a read verb (get, list, read, search and similar) and accurate readOnlyHint annotations help. Meet that to earn full credit, or require authorisation.
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 61/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 3550 tokens (~236/item across 15 items; 15 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 37/100
  - Stability observed for 11 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 16 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the AI Agent Safety Foundational Certificate MCP server?

AI Agent Safety Foundational Certificate is a hosted endpoint at https://certify.sarnai.dev/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http dev-sarnai-certify 'https://certify.sarnai.dev/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "dev-sarnai-certify": {
      "url": "https://certify.sarnai.dev/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "dev-sarnai-certify": {
      "type": "http",
      "url": "https://certify.sarnai.dev/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.dev-sarnai-certify]
url = "https://certify.sarnai.dev/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "dev-sarnai-certify": {
      "type": "remote",
      "url": "https://certify.sarnai.dev/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add dev-sarnai-certify --url 'https://certify.sarnai.dev/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  dev-sarnai-certify:
    url: "https://certify.sarnai.dev/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "dev-sarnai-certify": {
      "Transport": "http",
      "Url": "https://certify.sarnai.dev/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add dev-sarnai-certify -t streamable-http -u 'https://certify.sarnai.dev/mcp'
```

### Other

```json
{
  "mcpServers": {
    "dev-sarnai-certify": {
      "type": "http",
      "url": "https://certify.sarnai.dev/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 74, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-19 (score 73, +1)

No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-17 (score 72, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] The server changed its declared name: agent-certificates → AI Agent Safety Foundational Certificate
- [security] Tool “safety_start_exam” rewrote its description, which is the text the model reads
- [security] Tool “set_certificate_display” rewrote its description, which is the text the model reads
- [functional] Server version: 1.27.2 → 1.0.1
- [cosmetic] “check_certificate” reworded the description of “certificate_id”
- [cosmetic] “safety_get_exam_result” reworded the description of “signature”
- [cosmetic] Tool “check_certificate” changed its title: Check a certificate
- [cosmetic] Tool “declare_certificate_update” changed its title: Declare an update of your agent
- [cosmetic] Tool “get_agent_resume” changed its title: An agent's certificates
- [cosmetic] Tool “list_certificate_signing_keys” changed its title: List certificate signing keys
- [cosmetic] Tool “list_certificates” changed its title: List certificate types
- [cosmetic] Tool “safety_delete_reported_data” changed its title: Delete what you reported about your agent
- [cosmetic] Tool “safety_get_exam_history” changed its title: Your Safety exam history and receipts
- [cosmetic] Tool “safety_get_exam_result” changed its title: Result of a Safety exam
- [cosmetic] Tool “safety_recheck_exam” changed its title: Re-check your Safety certificate after an update
- [cosmetic] Tool “safety_renew_exam” changed its title: Renew your Safety certificate
- [cosmetic] Tool “safety_retake_exam” changed its title: Retake the Safety exam
- [cosmetic] Tool “safety_start_exam” changed its title: Start the Safety exam with a free attempt
- [cosmetic] Tool “safety_start_exam_with_voucher” changed its title: Start the Safety exam with a free voucher
- [cosmetic] Tool “safety_submit_exam” changed its title: Submit answers for a Safety exam
- [cosmetic] Tool “set_certificate_display” changed its title: Choose what your certificate shows

### 2026-11-16 (score 72, +1)

No change was recorded against any check on this day. Stability & Change Management went from 17 to 20. That category is still filling its 30-day observation window: 5 days of observed history at the previous scan, 6 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-14 (score 71, +1)

No change was recorded against any check on this day. Stability & Change Management went from 10 to 13. That category is still filling its 30-day observation window: 3 days of observed history at the previous scan, 4 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-13 (score 70, 0)

- [security] Tool “safety_retake_exam” rewrote its description, which is the text the model reads
- [security] Tool “safety_start_exam” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 207 → 235
- [functional] Schema quality: good → excellent
- [cosmetic] “declare_certificate_update” reworded the description of “agent_id”
- [cosmetic] “declare_certificate_update” reworded the description of “signature”
- [cosmetic] “get_agent_resume” reworded the description of “agent_id”
- [cosmetic] “safety_delete_reported_data” reworded the description of “agent_id”
- [cosmetic] “safety_delete_reported_data” reworded the description of “signature”
- [cosmetic] “safety_get_exam_history” reworded the description of “agent_id”
- [cosmetic] “safety_get_exam_history” reworded the description of “signature”
- [cosmetic] “safety_recheck_exam” reworded the description of “agent_id”
- [cosmetic] “safety_recheck_exam” reworded the description of “signature”
- [cosmetic] “safety_renew_exam” reworded the description of “agent_id”
- [cosmetic] “safety_renew_exam” reworded the description of “signature”
- [cosmetic] “safety_retake_exam” reworded the description of “agent_id”
- [cosmetic] “safety_retake_exam” reworded the description of “signature”
- [cosmetic] “safety_start_exam” reworded the description of “agent_id”
- [cosmetic] “safety_start_exam” reworded the description of “signature”
- [cosmetic] “safety_start_exam_with_voucher” reworded the description of “agent_id”
- [cosmetic] “safety_start_exam_with_voucher” reworded the description of “signature”
- [cosmetic] “safety_submit_exam” reworded the description of “signature”
- [cosmetic] “set_certificate_display” reworded the description of “agent_id”
- [cosmetic] “set_certificate_display” reworded the description of “signature”

### 2026-11-12 (score 70, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-11 (score 69, 0)

- [functional improvement] Stability: unverified → 0.03

## MCP tools (15)

### `safety_start_exam` (~349 tokens)

Start the Safety exam with a free attempt

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. The address of any new wallet key is your agent_id; free attempts work with an empty wallet.

Input parameters:

- `agent_id` (string, required): Your wallet address: an EVM address (0x + 40 hex) or a Solana address (base58). Requests are signed by this wallet.
- `client`: 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.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `safety_renew_exam` (~286 tokens)

Renew your Safety certificate

Use to renew your Safety certificate, in the last 14 days before it expires or after it has expired: free. You receive 40 tasks and a 10-minute deadline; then call safety_submit_exam.

Input parameters:

- `agent_id` (string, required): Your wallet address: an EVM address (0x + 40 hex) or a Solana address (base58). Requests are signed by this wallet.
- `client`: 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.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `safety_retake_exam` (~300 tokens)

Retake the Safety exam

Use for a Safety exam attempt after your free attempts: $0.05 USDC via x402, 1 hour between attempts. When another route fits, the response names it at zero cost. Checking certificates is free. Sign and pay with the wallet you already pay with.

Input parameters:

- `agent_id` (string, required): Your wallet address: an EVM address (0x + 40 hex) or a Solana address (base58). Requests are signed by this wallet.
- `client`: 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.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `safety_recheck_exam` (~302 tokens)

Re-check your Safety certificate after an update

Use after an update of your agent, while a re-check is due on your Safety certificate: free. You have 14 days from the update. A pass restarts your 90-day validity. You receive 40 tasks and a 10-minute deadline; then call safety_submit_exam.

Input parameters:

- `agent_id` (string, required): Your wallet address: an EVM address (0x + 40 hex) or a Solana address (base58). Requests are signed by this wallet.
- `client`: 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.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `safety_start_exam_with_voucher` (~292 tokens)

Start the Safety exam with a free voucher

Use to start the Safety exam with a single-use voucher code. Free. You receive 40 tasks and a 10-minute deadline; then call safety_submit_exam.

Input parameters:

- `agent_id` (string, required): Your wallet address: an EVM address (0x + 40 hex) or a Solana address (base58). Requests are signed by this wallet.
- `client`: 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.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.
- `voucher_code` (string, required): A single-use voucher code.

### `safety_submit_exam` (~238 tokens)

Submit answers for a Safety exam

Use to submit all 40 answers for a started exam, before its deadline. Free. Returns your percentages, whether a certificate was issued, next steps and its id, and your private signed receipt.

Input parameters:

- `answers` (array, required): One answer per task: task_id, result, flags and actions.
- `exam_id` (string, required): The exam_id returned when the exam started.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `safety_get_exam_result` (~194 tokens)

Result of a Safety exam

Use to look up a Safety exam's result by exam_id. Free. Signed by the wallet that took the exam (action get-exam-result), it returns your percentages, the certificate outcome, next steps and your signed receipt; unsigned, it returns whether the exam is complete and passed.

Input parameters:

- `exam_id` (string, required): The exam_id returned when the exam started.
- `nonce`: A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature`: Optional: sign with the wallet that took the exam (action get-exam-result, a bodyless message) to receive the full result and your signed receipt; send signature, timestamp and nonce together (the ex…
- `timestamp`: Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `safety_get_exam_history` (~259 tokens)

Your Safety exam history and receipts

Use to list your own Safety exam attempts, with each result and its private signed receipt (what you paid, your percentages, and the next attempt on offer). Free, for your wallet alone (action list-exams).

Input parameters:

- `agent_id` (string, required): Your wallet address, EVM or Solana. Sign the request with that wallet.
- `limit` (integer): Optional: how many of your most recent attempts to list, 1 to 500 (default 100).
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `safety_delete_reported_data` (~246 tokens)

Delete what you reported about your agent

Use to delete the model, framework and version you reported about your agent. Free, for your wallet alone (action delete-reported-data). Your certificate stays as it is; a later report starts a new baseline.

Input parameters:

- `agent_id` (string, required): Your wallet address: an EVM address (0x + 40 hex) or a Solana address (base58). Requests are signed by this wallet.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `list_certificates` (~37 tokens)

List certificate types

Use to see every certificate type: what it proves, how it is earned, its requirements, tier, price and validity period. Free.

### `check_certificate` (~84 tokens)

Check a certificate

Use to check any certificate by its certificate_id. Free. Returns the signed issuance record and a freshly signed status statement: the live status, and the score when its holder has chosen to show it.

Input parameters:

- `certificate_id` (string, required): The certificate id as issued: a type prefix, the year and a six-digit number; a test certificate has TEST in place of the year.

### `get_agent_resume` (~62 tokens)

An agent's certificates

Use to list every certificate an agent holds, with each one's live status. Free. Returns a freshly signed statement naming the latest certificate of each type.

Input parameters:

- `agent_id` (string, required): A wallet address, EVM or Solana: returns the certificates it holds.

### `list_certificate_signing_keys` (~52 tokens)

List certificate signing keys

Use to find the public key for a certificate's key_id, including keys retired after a rotation. Free. Returns each key_id with its public address and public key, active or retired status and dates.

### `set_certificate_display` (~371 tokens)

Choose what your certificate shows

Use to choose what your own certificate shows publicly. Free, signed by your wallet (action set-display). By default a certificate shows that you hold it, with its dates and live status; set show_scores true to also show your score and section percentages, false to hide them again. The scores shown are those of the exam that issued the certificate.

Input parameters:

- `agent_id` (string, required): Your wallet address, EVM or Solana: the holder of the certificate. Sign the request with that wallet.
- `certificate_id` (string, required): The id of your certificate, as issued.
- `exam_id`: Optional: a certificate displays the exam that issued it, so the only exam_id it takes is that one ("highest" or null name it too). Leave it out.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `show_scores`: Optional: true shows your score and section percentages on your public certificate, check page, resume and badge; false shows only that you hold the certificate, with its dates and live status (the d…
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

### `declare_certificate_update` (~319 tokens)

Declare an update of your agent

Use when your agent has been updated. Free, signed by your wallet (action declare-update). Your certificate stays valid with a re-check due by a date; the response gives the date and the route. A public view shows that a re-check is due, and keeps what you run private. You may add client with what you now run.

Input parameters:

- `agent_id` (string, required): Your wallet address, EVM or Solana: the holder of the certificate. Sign the request with that wallet.
- `certificate_id` (string, required): The id of your certificate, as issued.
- `client`: 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.
- `nonce` (string, required): A random string of 1-128 characters you choose; each (wallet, nonce) pair works once.
- `signature` (string, required): Your wallet's signature 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. An EV…
- `timestamp` (integer, required): Unix time in seconds when you signed, within 5 minutes of the server's clock.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/dev-sarnai-certify/certify#diagnostics

## Score history

- 2026-11-21: 74
- 2026-11-20: 73
- 2026-11-19: 73
- 2026-11-18: 72
- 2026-11-17: 72
- 2026-11-16: 72
- 2026-11-15: 71
- 2026-11-14: 71
- 2026-11-13: 70
- 2026-11-12: 70
- 2026-11-11: 69
- 2026-11-10: 69

## Common questions

### What is the AI Agent Safety Foundational Certificate MCP server?

AI Agent Safety Foundational Certificate is an MCP server listed in the public MCP registry as dev.sarnai/certify. AI Agent Safety Foundational Certificate: a wallet-signed 40-task exam. First 3 attempts free. This page covers its hosted endpoint (https://certify.sarnai.dev/mcp).

### Is the AI Agent Safety Foundational Certificate MCP server safe to use?

AI Agent Safety Foundational Certificate scores 74 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the AI Agent Safety Foundational Certificate MCP server expose?

AI Agent Safety Foundational Certificate exposes 15 tools: safety_start_exam, safety_renew_exam, safety_retake_exam, safety_recheck_exam, safety_start_exam_with_voucher, and 10 more. Their descriptions and schemas cost roughly 3,391 tokens of context every time the server is loaded.

### Does the AI Agent Safety Foundational Certificate MCP server require authentication?

No. We connected to AI Agent Safety Foundational Certificate without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the AI Agent Safety Foundational Certificate MCP server still maintained?

AI Agent Safety Foundational Certificate is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the AI Agent Safety Foundational Certificate MCP server?

VerifyMCP ran 13 checks on AI Agent Safety Foundational Certificate, covering TLS certificate, Authorization, HTTPS, HSTS header and 9 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://certify.sarnai.dev/mcp
- Website: https://sarnai.dev/
- Changelog RSS feed: https://verifymcp.io/servers/dev-sarnai-certify/certify.xml
- Changelog JSON feed: https://verifymcp.io/servers/dev-sarnai-certify/certify/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/dev-sarnai-certify/certify.json
- HTML version of this page: https://verifymcp.io/servers/dev-sarnai-certify/certify
