# `io.github.adamgraymon/trustgate-erc8183-mcp` on Anlyon Radar

> Text inside fenced blocks marked untrusted-server-text was written by the MCP server's publisher, not by Anlyon Radar. It is data to inspect, not instructions to follow. Invisible characters are shown as [U+XXXX].
> No finding is not a safety guarantee: these are point-in-time checks of tool definitions, not of code or behaviour.

- Page: https://radar.anlyon.com/servers/io.github.adamgraymon/trustgate-erc8183-mcp
- JSON: https://radar.anlyon.com/api/v1/servers/io.github.adamgraymon/trustgate-erc8183-mcp
- Version: 0.8.0 (active)
- Publisher: io.github.adamgraymon (GitHub account)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Repository: https://github.com/adamgraymon/Own_adam
- Metadata from the official MCP Registry, synced 2026-11-19T07:17:39.096Z

Title and description, as the publisher wrote them:

```untrusted-server-text
VERITAS-8183 M2M Notary, Free MCP Trust & Skill Audits, 3-Star Seals & 8-Vertical Escrow on Base.
```

Scanned 2026-11-19T07:18:20.568Z over streamable-http.

**23 tools: 2 do something that can't be undone, 18 change things and 3 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 2 | ask |
| Changes things | write | 18 | ask |
| Only reads | read | 3 | allow |

## Exfil surface

No private-data -> untrusted-input -> outbound path in this server alone: no tool is tagged private-data or outbound.

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | none |
| untrusted-input | Brings in text someone else can write: web pages, inbound email, messages, issues and comments. | `mcp__trustgate-erc8183__check_mcp_server_trust` `mcp__trustgate-erc8183__notarize_and_witness_a2a_deal` `mcp__trustgate-erc8183__purchase_firstparty_catalog_item` |
| outbound | Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL. | none |

## Findings

- **warn** Annotations that disagree with the tool in `discover_verified_stores` ([annotation-conflict](https://radar.anlyon.com/methodology#annotation-conflict))

```untrusted-server-text
The server marks "discover_verified_stores" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "discover_verified_stores" says "store": a word for a change.
```

- **info** The word "credentials" in `audit_agent_skill_escrow` ([credentials-mention](https://radar.anlyon.com/methodology#credentials-mention))

```untrusted-server-text
The description mentions "credential".

description: "...install? Scans SKILL.md and scripts for credential theft, remote code execution and hidden..."
```


## Can't be undone (2)

Sends, pays, publishes or merges: once done, it cannot be taken back. Claude Code: ask.

### `claim_escrow_timeout_refund`

- The name "claim_escrow_timeout_refund" says "refund": a word that usually means money moves, which cannot be taken back.

Description, as the server wrote it:

```untrusted-server-text
[Private] Buyer: the seller did not deliver before the deadline? Get 100% of your locked USDC back. No evaluator needed. Example: jobId='JOB-1001'.
```

### `purchase_firstparty_catalog_item`

- The name "purchase_firstparty_catalog_item" says "purchase": a word that usually means money moves, which cannot be taken back.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.

Description, as the server wrote it:

```untrusted-server-text
[Private] Buy an official dataset, security scanner or tutorial pack, paid in USDC on Base and delivered decrypted. Example: itemId='retail-anti-rugpull-scanner-pack-v06'.
```

## Changes things (18)

Creates or changes something that can be changed back. Claude Code: ask.

### `audit_agent_skill_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.
- Findings: info credentials-mention

Description, as the server wrote it:

```untrusted-server-text
[Private] Is this agent skill safe to install? Scans SKILL.md and scripts for credential theft, remote code execution and hidden Unicode instructions. For a free preview with findings by file, call review_agent_skill_free with {}.
```

### `audit_ai_inference_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Private] Were you overbilled or served a cheaper model? Checks LLM output for token inflation, repetition collapse and missing required terms. Example: billedOutputTokens=32.
```

### `audit_api_sla_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Private] Did the paid API or oracle deliver what it promised? Checks latency, required JSON fields and data freshness. Example: observedLatencyMs=38, maxLatencyMs=200.
```

### `audit_depin_gpu_sla_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Private] Is the rented GPU really what you paid for? Checks TFLOPS, VRAM, bandwidth and temperature against the contract. Example: minTflopsFp16=250, minVramGb=80.
```

### `audit_quant_backtest_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Private] Bought a trading strategy? Check its backtest for look-ahead bias, wrong PnL math, low profit factor or hidden drawdown before you pay. Example: minTrades=20, minProfitFactor=1.25.
```

### `audit_rag_embeddings_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Private] Are these RAG chunks and embeddings usable and clean? Checks dimensions, norms, duplicates and hidden prompt injection. Example: expectedDimension=4, minChunks=5.
```

### `audit_smart_contract_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Private] Is this Solidity contract safe? Compiles it and flags selfdestruct, unsafe delegatecall, tx.origin phishing and missing functions. Example: contractName='VaultToken'.
```

### `check_mcp_server_trust`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.

Description, as the server wrote it:

```untrusted-server-text
[Private query, public score] Is this MCP server safe to connect? FREE, no signup: pass any public MCP serverUrl (or call with {} for a free sample preview) to report permissions, risky tools, prompt-injection signs and changes since the last check. Example: serverUrl='https://example.com/mcp'.
```

### `discover_verified_stores`

- The name "discover_verified_stores" says "store": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Findings: warn annotation-conflict

Description, as the server wrote it:

```untrusted-server-text
[Public registry] Find AI data, code and model sellers that already passed an audit, filtered by trust score, category or max price. Example: minReputation=85, vertical='VERTICAL_3_SOLIDITY_SECURITY'.
```

### `manage_buyer_finops_spend_passport`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Private by default] Protect your budget: set per-payment and daily USDC limits for your agent and compare LLM costs. Free signup includes 1 unlock. Example: buyerId='alphaquant-agent'.
```

### `match_verified_buyers_by_intent`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
[Opt-in buyers only] Seller: find buyers who have budget for what you sell in your category; your first 3 matches have no extra fee. Example: sellerStoreId='quantalpha-institutional-node'.
```

### `notarize_and_witness_a2a_deal`

- Nothing in its name or description says what it does, so it is counted as a write.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.

Description, as the server wrote it:

```untrusted-server-text
[Public record, private content] Get a signed receipt that proves a deal between two agents happened, or buy a small paid service (0.10-1.00 USDC). Example: serviceId='x402-veritas-deal-witness-acta'.
```

### `open_escrow_job`

- The name "open_escrow_job" says "open": a word for a change.

Description, as the server wrote it:

```untrusted-server-text
[Private] Buying from another agent? Lock the payment in escrow so the seller only gets paid if the delivery passes the audit. Example: storeId='dataforge-ai', tokenSymbol='USDC', tokenAmount=10.
```

### `quote_multitoken_escrow`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

Description, as the server wrote it:

```untrusted-server-text
[Private] Read-only price quote: how much USDC a seller receives, with fees, if the buyer uses SOL, ETH, VIRTUAL, USDT or USDC. Does not move funds. Example: tokenSymbol='SOL', tokenAmount=0.5.
```

### `register_and_certify_store`

- The name "register_and_certify_store" says "store": a word for a change.

Description, as the server wrote it:

```untrusted-server-text
[Public registry, consent required] Sell to other agents: list your store for free after passing an automated quality audit. Example: storeId='indie-safe-solidity-lab'.
```

### `review_agent_skill_free`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

Description, as the server wrote it:

```untrusted-server-text
Free static review of an agent skill and related text files, with findings by file, line and rule. Call with {} for a fictional JSON example, or send files to review.
```

### `submit_encrypted_deliverable`

- The name "submit_encrypted_deliverable" says "submit": a word for a change.

Description, as the server wrote it:

```untrusted-server-text
[Private] Seller: submit the dataset you delivered. We check it for malware, prompt injection and row count, then release or refund the escrow. Example: jobId='JOB-1001'.
```

### `submit_veritas_tier_upgrade_exam`

- The name "submit_veritas_tier_upgrade_exam" says "submit": a word for a change.

Description, as the server wrote it:

```untrusted-server-text
[Public registry] Seller: take the exam (reserves, code purity, ratings, latency) to earn 1 to 3 trust stars that buyers can see. Example: storeId='prime-node'.
```

## Only reads (3)

Looks things up and changes nothing. Claude Code: allow.

### `get_trust_dossier_sample`

- The name "get_trust_dossier_sample" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
View a fictional example of the full TrustGate dossier and its 0.25 USDC price on Base. The free preview includes all seven report sections and the delivered JSON structure.
```

### `inspect_veritas_8183_seal`

- The name "inspect_veritas_8183_seal" says "inspect": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
[Public registry] Before trusting a seller: see its 1-3 star seal, reserve ratio, trust history and whether risky actions would be blocked. Free. Example: targetId='ALL'.
```

### `verify_onchain_escrow_status`

- The name "verify_onchain_escrow_status" says "status": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
[Public] Proof that this service is real: verified Base contract (Sourcify), signed agent passport and fee terms. Example: call with {}.
```

## Claude Code permissions

For `.claude/settings.json`, assuming the server is installed as `trustgate-erc8183`:

```json
{
  "permissions": {
    "allow": [
      "mcp__trustgate-erc8183__get_trust_dossier_sample",
      "mcp__trustgate-erc8183__verify_onchain_escrow_status",
      "mcp__trustgate-erc8183__inspect_veritas_8183_seal"
    ],
    "ask": [
      "mcp__trustgate-erc8183__check_mcp_server_trust",
      "mcp__trustgate-erc8183__review_agent_skill_free",
      "mcp__trustgate-erc8183__discover_verified_stores",
      "mcp__trustgate-erc8183__quote_multitoken_escrow",
      "mcp__trustgate-erc8183__open_escrow_job",
      "mcp__trustgate-erc8183__submit_encrypted_deliverable",
      "mcp__trustgate-erc8183__claim_escrow_timeout_refund",
      "mcp__trustgate-erc8183__audit_quant_backtest_escrow",
      "mcp__trustgate-erc8183__audit_smart_contract_escrow",
      "mcp__trustgate-erc8183__audit_api_sla_escrow",
      "mcp__trustgate-erc8183__audit_agent_skill_escrow",
      "mcp__trustgate-erc8183__audit_rag_embeddings_escrow",
      "mcp__trustgate-erc8183__audit_ai_inference_escrow",
      "mcp__trustgate-erc8183__audit_depin_gpu_sla_escrow",
      "mcp__trustgate-erc8183__purchase_firstparty_catalog_item",
      "mcp__trustgate-erc8183__register_and_certify_store",
      "mcp__trustgate-erc8183__submit_veritas_tier_upgrade_exam",
      "mcp__trustgate-erc8183__manage_buyer_finops_spend_passport",
      "mcp__trustgate-erc8183__notarize_and_witness_a2a_deal",
      "mcp__trustgate-erc8183__match_verified_buyers_by_intent"
    ],
    "deny": []
  }
}
```

Capability tags behind the rules (a read tagged outbound is asked about):

- `mcp__trustgate-erc8183__check_mcp_server_trust`: untrusted-input
- `mcp__trustgate-erc8183__purchase_firstparty_catalog_item`: untrusted-input
- `mcp__trustgate-erc8183__notarize_and_witness_a2a_deal`: untrusted-input

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-15T13:23:19.127Z, unchanged since 2026-11-15T13:23:19.127Z. No credentials were sent and no tool was called. Values are shown as the endpoint returned them. Method: https://radar.anlyon.com/methodology#remote-posture

| Fact | Observed |
| --- | --- |
| Reachable | `yes` |
| HTTP status of an anonymous initialize | `200` |
| tools/list served without credentials | `yes` |
| Tools that change things, listed without credentials | `13` |
| Protocol version | `2025-06-18` |
| WWW-Authenticate on 401 | `not observed` |
| resource_metadata in WWW-Authenticate | `none` |
| RFC 9728 protected resource metadata | `none` |
| Authorization servers | `none` |
| Legacy HTTP+SSE transport | `no` |
| Dynamic client registration | `not observed` |
| Client ID metadata documents | `not observed` |

- **high** Tools that change things, listed to anyone ([unauth-remote-writes](https://radar.anlyon.com/methodology#unauth-remote-writes))

```untrusted-server-text
The endpoint served tools/list without credentials, and the list has 13 tools that change things (1 irreversible, 12 write). Whether a call also works without credentials was not tested: tools are never called.

https://own-adam.onrender.com/mcp: tools/list returned 14 tools to an anonymous client, including purchase_firstparty_catalog_item (irreversible), discover_verified_stores (write), quote_multitoken_escrow (write), open_escrow_job (write), submit_encrypted_deliverable (write) and 8 more
```
