# FHI MCP Server Security Audit

> Free payment guide and Base-USDC x402 MCP security, package, domain, and SEC tools.

- name: dev.workers.fhi-llc-1118.polished-truth-c514/fhi-x402-security-tools
- version: 1.4.0
- connection class: R0 (autonomous)
- trust: 80/100 flags: no-repository
- quality: 79/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/dev.workers.fhi-llc-1118.polished-truth-c514%2Ffhi-x402-security-tools
- tools: https://api.protogrid.dev/v1/servers/dev.workers.fhi-llc-1118.polished-truth-c514%2Ffhi-x402-security-tools/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "fhi-x402-security-tools": {
      "type": "http",
      "url": "https://polished-truth-c514.fhi-llc-1118.workers.dev/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 100
- freshness: 100
- provenance: 45
- drivers: +dns-namespace +reachable +uptime-100% +updated-18d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 79/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 71 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- fail `tools.annotations`: no tool declares readOnlyHint or destructiveHint, so clients cannot tell safe calls from risky ones
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: mcp_payment_preflight (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); mcp_vendor_due_diligence (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); payment_info (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); …
- pass `tools.token_cost`: about 783 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: 3 tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 7, about 783 tokens to load them all
- tool set hash: c7069cf12168e51c0d491eaa63fab6ad
- tools last changed: 2026-10-07T13:46:39.226Z
- badge: [![protogrid quality](https://protogrid.dev/badge/dev.workers.fhi-llc-1118.polished-truth-c514/fhi-x402-security-tools.svg)](https://protogrid.dev/servers/dev.workers.fhi-llc-1118.polished-truth-c514/fhi-x402-security-tools)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-07 `mcp_change_evidence` changed (description)
  - before: Stateful live remote-MCP monitor for tool-rug-pull and tool-poisoning risk. Establishes an endpoint baseline, then detects added or removed tools, altered tool descriptions, risk-score changes, and newly observed high-risk capability signals before an agent reconnects or pays. Unlike a manifest diff, it fetches the live endpoint and retains the prior observation. ($0.01 USDC on Base)
  - after: Stateful live remote-MCP monitor for tool-rug-pull and tool-poisoning risk. Establishes an endpoint baseline, then detects added or removed tools, altered tool descriptions, risk-score changes, and newly observed high-risk capability signals before an agent reconnects or pays. Unlike a manifest diff, it fetches the live endpoint and retains the prior observation. ($0.003 USDC on Base)
- 2026-10-07 `package_install_preflight` changed (description)
  - before: npm or PyPI install preflight: blocks missing packages, detects near-name typosquats when an expected name is supplied, and checks OSV advisories. ($0.01 USDC on Base)
  - after: npm or PyPI install preflight: blocks missing packages, detects near-name typosquats when an expected name is supplied, and checks OSV advisories. ($0.003 USDC on Base)
- 2026-10-06 `mcp_change_evidence` added
  - Stateful live remote-MCP monitor for tool-rug-pull and tool-poisoning risk. Establishes an endpoint baseline, then detects added or removed tools, altered tool descriptions, risk-score changes, and newly observed high-risk capability signals before an agent reconnects or pays. Unlike a manifest diff, it fetches the live endpoint and retains the prior observation. ($0.01 USDC on Base)

Full history: https://api.protogrid.dev/v1/servers/dev.workers.fhi-llc-1118.polished-truth-c514%2Ffhi-x402-security-tools/changes

## Remotes

- https://polished-truth-c514.fhi-llc-1118.workers.dev/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (7)

- `domain_change_evidence`: Stateful DNS, CAA, and certificate-transparency monitoring for a public domain. The first call establishes a baseline; later calls return evidence changes and certificate-expiry signals. ($0.01 USDC on Base)
- `mcp_change_evidence`: Stateful live remote-MCP monitor for tool-rug-pull and tool-poisoning risk. Establishes an endpoint baseline, then detects added or removed tools, altered tool descriptions, risk-score changes, and newly observed high-risk capability signals before an agent reconnects or pays. Unlike a manifest diff, it fetches the live endpoint and retains the prior observation. ($0.003 USDC on Base)
- `mcp_payment_preflight`: MCP server security audit before an agent connects or pays: probes initialize and tools/list, then returns an ALLOW, CAUTION, or BLOCK risk score for command, filesystem, or wallet capabilities; prompt-injection or data-exfiltration signals; permissive JSON-schema inputs; missing HSTS; and a large tool surface. Does not execute tools. ($0.01 USDC on Base)
- `mcp_vendor_due_diligence`: One decision-ready due-diligence dossier before an agent adopts or pays an MCP vendor: live MCP metadata and tool-risk preflight, DNS/CAA/certificate evidence, plus optional SEC material-filing evidence for a public company. Does not execute vendor tools or certify safety. ($0.10 USDC on Base)
- `package_install_preflight`: npm or PyPI install preflight: blocks missing packages, detects near-name typosquats when an expected name is supplied, and checks OSV advisories. ($0.003 USDC on Base)
- `payment_info`: Free guide to the FHI MCP tools, Base-USDC x402 payment flow, and per-tool prices.
- `sec_material_event_delta`: New SEC 8-K, 6-K, and late-filing evidence since a cursor date; accepts a ticker or CIK. ($0.01 USDC on Base)