# Plainport

> Deterministic web intake and data utilities for autonomous agents.

- name: dev.workers.exportitnow.plainport/plainport
- version: 0.1.0
- connection class: R0 (autonomous)
- trust: 79/100 flags: no-repository
- quality: 100/100 (strong)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/dev.workers.exportitnow.plainport%2Fplainport
- tools: https://api.protogrid.dev/v1/servers/dev.workers.exportitnow.plainport%2Fplainport/tools

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

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "plainport": {
      "type": "http",
      "url": "https://plainport.exportitnow.workers.dev/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 99
- freshness: 85
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-98% ~updated-46d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 100/100 (strong)

### protocol: 100 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- pass `protocol.list_ttl`: tool list is cacheable (ttlMs 300000)

### 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`: 10 of 10 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 1,887 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: 10, about 1,887 tokens to load them all
- tool set hash: 1e6b1b30e253a7a226594d165c5dc971
- badge: [![protogrid quality](https://protogrid.dev/badge/dev.workers.exportitnow.plainport/plainport.svg)](https://protogrid.dev/servers/dev.workers.exportitnow.plainport/plainport)
- 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://plainport.exportitnow.workers.dev/mcp (streamable-http, auth none, reachable true, uptime30d 0.9830508)

## Packages

_none_

## Tools (10)

- `diff_text`: Computes a deterministic line- or word-level diff and returns typed add/remove/equal parts. Use this when an agent needs structured change data. Split highly divergent large documents if the computation limit is reached. Limit: 200k combined characters, 2,000 result parts, 8 ms diff budget.
- `extract_links`: Fetches one public HTML page and returns unique normalized HTTP/HTTPS links with anchor text, rel values, and same-origin classification. Use this for bounded navigation discovery; it does not crawl the returned links. Limit: 8 s, 5 redirects, 768 KiB, 500 links.
- `extract_metadata`: Fetches one public HTML page and extracts title, description, canonical URL, robots directives, Open Graph, Twitter Card, advertised feeds, and parsed JSON-LD. Use this when an agent needs page identity or structured data rather than body text. Limit: 8 s, 5 redirects, 768 KiB, 50 JSON-LD blocks.
- `fetch_markdown`: Fetches one public HTTP/HTTPS page and returns its primary readable content as Markdown. Use this for article or documentation body text without page chrome. It does not run JavaScript; use a browser tool for client-rendered pages. Native text/markdown responses are preserved. Limit: 8 s, 5 redirects, 768 KiB remote body, 150k output characters.
- `html_to_markdown`: Converts supplied HTML into readable Markdown without fetching a URL. Use this when the agent already has HTML. Provide base_url only to resolve relative links; use fetch_markdown when Plainport should retrieve the page. Limit: 200k input characters, 150k output characters.
- `inspect_robots`: Fetches an origin's robots.txt and evaluates one path for a named user agent using longest-match allow/disallow precedence. Also returns crawl delay and declared sitemaps. Use this before automated retrieval; the result is policy guidance, not authorization. Limit: 8 s, 5 redirects, 768 KiB, 50 sitemap declarations.
- `inspect_url`: Resolves a public URL through bounded redirects and returns status plus a safe allowlist of HTTP headers. Use HEAD for cheap reachability/type checks and GET when a server mishandles HEAD. Use extract_metadata for HTML semantic metadata. Limit: 8 s, 5 redirects, 768 KiB only for GET.
- `parse_feed`: Fetches and normalizes RSS 2.0, Atom 1.0, or JSON Feed into a stable item list. Use this when consuming update feeds without format-specific parsing. It does not fetch linked articles. Limit: 8 s, 5 redirects, 768 KiB, 100 items.
- `parse_sitemap`: Fetches one uncompressed XML sitemap and returns either URL entries or child sitemap references. Use this for site structure discovery. Sitemap indexes are not recursively expanded; call this tool on each child explicitly. Limit: 8 s, 5 redirects, 768 KiB, 1,000 entries.
- `validate_json_schema`: Validates a JSON value against a supplied JSON Schema using draft 4, 7, 2019-09, or 2020-12. Use this to verify generated or extracted structured data. It performs validation only and does not mutate the instance. Limit: 512 KiB request, 40 nesting levels, 100 returned errors.