# Nanotoll MCP

> 7 agent tools (prune, convert, estimate, diff, patch, generate, validate). USDC on Base L2.

- name: dev.nanotoll/nanotoll-mcp
- version: 0.1.0
- connection class: R1 (autonomous)
- trust: 80/100 flags: no-repository
- quality: 96/100 (strong)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/dev.nanotoll%2Fnanotoll-mcp
- tools: https://api.protogrid.dev/v1/servers/dev.nanotoll%2Fnanotoll-mcp/tools

**An agent can connect on its own if it already holds `apiKey`.** Remote endpoint behind a static API key or header. The registry never sees the value; the agent fills the placeholder.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "nanotoll-mcp": {
      "type": "http",
      "url": "https://mcp.nanotoll.dev/mcp",
      "headers": {
        "Authorization": "Bearer ${apiKey}"
      }
    }
  }
}
```

Placeholders to fill: `${apiKey}`.

## Trust

- hygiene: 80
- liveness: 100
- freshness: 85
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-100% ~updated-60d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 96/100 (strong)

### protocol: 88 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- warn `protocol.list_ttl`: tools/list carries no ttlMs, so clients re-fetch it

### 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: 100 (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
- pass `tools.annotations`: 8 of 8 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 1,841 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`: no 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: 8, about 1,841 tokens to load them all
- tool set hash: ec06d4d4a9c174ebe9bb63b937bd16b6
- badge: [![protogrid quality](https://protogrid.dev/badge/dev.nanotoll/nanotoll-mcp.svg)](https://protogrid.dev/servers/dev.nanotoll/nanotoll-mcp)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

_No tool definition changes recorded._

## Remotes

- https://mcp.nanotoll.dev/mcp (streamable-http, auth api_key ${apiKey}, reachable true, uptime30d 1)

## Packages

_none_

## Tools (8)

- `convert`: Transform text between JSON, YAML, TOML, and CSV formats. Returns the converted text and input/output character metrics. Note some conversions can be lossy (e.g. JSON to CSV flattens nested structures). Requires a valid API key (Bearer token); billing is per input character — insufficient balance returns HTTP 402. For removing comments use `prune`, for generating llms.txt use `generate`.
- `diff`: Generate a unified diff between two text versions (the before and after). Returns the diff text plus metrics (lines added, lines removed, output chars). Requires a valid API key (Bearer token); billing is per combined input characters. To apply a diff, use `patch`.
- `estimate`: Estimate the USD cost of an LLM API call using the CWEP four-flow model (request generation, request processing, response generation, response reception). Returns a JSON object with the per-flow cost breakdown, the total, and the pricing source date. Unknown model names return HTTP 422 — check GET /api/v1/pricing on costa.nanotoll.dev for the supported list (no auth needed). Requires a valid API key (Bearer token); billing is per input character.
- `generate`: Generate an llms.txt agent-discovery file from structured service metadata (name, endpoints, overview, links, pricing). Returns the llms.txt content and output metrics. Requires a valid API key (Bearer token); billing is per input character. To check an existing llms.txt for spec compliance, use `validate`.
- `patch`: Apply a unified diff patch to text, producing the transformed output. Returns the patched text. Requires a valid API key (Bearer token); billing is per combined input characters. To generate a diff, use `diff`.
- `prune`: Strip comments and empty lines from text to reduce token count before sending to an LLM. Returns the pruned text and character metrics. Requires a valid API key (Bearer token); billing is per input character — insufficient balance returns HTTP 402. For format conversion use `convert`, for generating llms.txt use `generate`.
- `read`: Fetch a web page and return clean GitHub-flavored markdown at a strict token budget. DOM-density scoring strips boilerplate (CSS, JS, navbars, tracking) — raw HTML averages 80-95% noise. Requires a valid API key (Bearer token); billing is $0.0005 flat per read — insufficient balance returns HTTP 402. Fetch failures return HTTP 422 and are refunded.
- `validate`: Validate an llms.txt payload for spec compliance (H1 title, endpoints section, pricing section, structure). Returns a validation report with errors, warnings, and structural flags. Requires a valid API key (Bearer token); billing is per input character. To generate new llms.txt, use `generate`.