Gopher & Gemini MCP Server

PyPI version Downloads

A modern, cross-platform Model Context Protocol (MCP) server that enables AI assistants to browse and interact with both Gopher protocol and Gemini protocol resources safely and efficiently.

Overview

The Gopher & Gemini MCP Server bridges vintage and modern alternative internet protocols with AI assistants, allowing LLMs like Claude to explore the unique content and communities that thrive on both Gopherspace and Geminispace. Built with FastMCP and modern Python practices, it provides secure, efficient gateways to these distinctive internet protocols.

Key Benefits:

• Discover alternative internet content - Access unique resources on both Gopher and Gemini protocols
• Safe exploration - Built-in security safeguards, TLS encryption, and content filtering
• Modern implementation - Uses FastMCP framework with async/await patterns
• Developer-friendly - Comprehensive testing, type hints, and documentation
• Advanced security - TOFU certificate validation and client certificate support for Gemini

Features

• Dual Protocol Support: gopher_fetch and gemini_fetch tools for comprehensive protocol coverage
• Comprehensive Gopher Support: Every standard RFC 1436 item type — menus (1) and Index-Search servers (7) as structured menus, text (0), HTML (h), info (i) and error (3) lines as text, the fourteen binary types as metadata only, and the three interactive ones (2, 8, T) refused without opening a connection. The one standard type with no category of its own is + (redundant server), which names an alternate host for the preceding item rather than content to render; it takes the unknown-type path below. An unknown type is read as text, best-effort, and an hURL URL:<target> selector is followed to the destination the server actually stated
• Full Gemini Implementation: Native gemtext parsing, TLS security, and status code handling
• Advanced Security: TOFU certificate validation with dedicated inspection and recovery tools, scoped client certificates, and secure TLS connections
• Safety First: Built-in timeouts, size limits, input sanitization, SSRF protection, per-host rate limiting, and host allowlists
• LLM-Optimized: Returns structured JSON responses designed for AI consumption
• Cross-Platform: Works seamlessly on Windows, macOS, and Linux
• Modern Development: Full type checking, linting, testing, and CI/CD pipeline
• High Performance: Async/await patterns with intelligent caching — and cached results say so, with a per-request refresh bypass
• Continuable Reads: A menu or page cut at the render limit reports where it stops, so offset reads the rest instead of leaving a partial view

Protocols in scope

Gopher (RFC 1436, including the de-facto item types in common use) and Gemini, both read-only. That is the whole surface, and the neighbouring protocols are deliberately out of scope rather than merely unbuilt:

• Titan and Misfin are write protocols — upload and mail. A tool an LLM drives should not be able to publish to someone's capsule or send mail as them, and adding either would make every safeguard here (robots, rate limits, allowlists) protect a much smaller share of what the tool can do.
• Spartan and Nex are separate protocols with their own parsers and their own failure modes, serving a small fraction of the hosts these two do. They would double the security-relevant surface for a rounding error in reach.
• Gopher+ is not implemented. The : item type is recognised because it appears in ordinary menus, but no Gopher+ attribute or metadata request is ever sent.

If you need one of these, an MCP server that does it well is a better answer than a flag on this one.

Documentation

Complete documentation is available at cameronrye.github.io/gopher-mcp

• Installation Guide
• Configuration Guide
• API Reference
• AI Assistant Guide
• Migration Guide and Changelog — what changed, and what an upgrade asks of you

Quick Start

Prerequisites

• Python 3.11+ - Download here
• uv package manager - Install uv

Installation

Option 1: Zero-install with uvx (Recommended)

No clone, no checkout — uv fetches and runs the published package on demand:

uvx gopher-mcp

Option 2: PyPI Installation

# Install from PyPI
pip install gopher-mcp

# Or with uv
uv add gopher-mcp

Option 3: Development Installation

# Clone the repository
git clone https://github.com/cameronrye/gopher-mcp.git
cd gopher-mcp

# Set up development environment
./scripts/dev-setup.sh  # Unix/macOS
# or
scripts\dev-setup.bat   # Windows

# Run the server
uv run task serve

Option 4: Docker

Tagged releases publish a slim, non-root image to ghcr.io/cameronrye/gopher-mcp, tagged with the release version plus :latest for stable (non-pre-release) tags:

# The default CMD serves streamable-http on 0.0.0.0:8000
docker run --rm -p 8000:8000 \
  -v gopher-mcp-state:/home/app/.local/share/gopher-mcp \
  ghcr.io/cameronrye/gopher-mcp:latest

# Or run over stdio, e.g. for an MCP client
docker run --rm -i --no-healthcheck \
  -v gopher-mcp-state:/home/app/.local/share/gopher-mcp \
  ghcr.io/cameronrye/gopher-mcp:latest --transport stdio

To run a modified tree, the repository ships the Dockerfile the published image is built from: docker build -t gopher-mcp .

Mount a volume, or Gemini trust is meaningless. Without one, the TOFU pins and the client certificates' private keys die with the container, so every start re-arms blind trust-on-first-use — the pin is the only thing that authenticates a Gemini capsule — and destroys any identity you minted, whose private key cannot be recovered.

Mount it at that exact path. /home/app/.local/share/gopher-mcp is where the server writes (tofu.json and certs/), and it is the one directory the image pre-creates owned by the runtime user and mode 700 — which is what lets a named volume come up writable instead of root-owned. Mounting anywhere else persists an empty directory. ~/.gemini is not the path: it is only a read-in-place upgrade route for installs that pinned certificates before gopher-mcp had a directory of its own, and it is honoured only when its store file is already there, which it never is in a fresh image.

Health checks. The HTTP transports serve GET /health, which answers {"status": "ok", "version": "..."} and nothing else — no configuration, no allowlists, no store paths. It bypasses authorization by SDK design, which is what makes it usable as a probe. The image's HEALTHCHECK polls it on the hard-coded port 8000 to match the default CMD, so override the healthcheck alongside --port, and pass --no-healthcheck when running stdio — a stdio container serves no HTTP and would otherwise be reported unhealthy while working perfectly.

Note: the default CMD binds 0.0.0.0 so the container is reachable out of the box. A non-loopback --host also turns off FastMCP's DNS-rebinding Host/Origin check, matching what the SDK does when it is constructed with such a host — otherwise every client that was not on localhost got 421 Misdirected Request. Keep the check on by naming the hostnames the deployment answers to with --allowed-host NAME (repeatable; a bare name matches any port). The HTTP transports are unauthenticated and have no TLS either — put the container behind a trusted reverse proxy, or use --transport stdio, before exposing it beyond your machine.

MCP Client Integration

Every client below runs the server over stdio — no ports, no TLS, no listening socket. The entry is the same three fields everywhere; only the file and the top-level key change:

{
  "mcpServers": {
    "gopher": {
      "command": "uvx",
      "args": ["gopher-mcp"]
    }
  }
}

That is the whole entry: every setting in Configuration has a working default, so add an "env" block only when you actually want to change one. Installed with pip rather than uvx? Use "command": "gopher-mcp" and "args": [] instead.

The Installation Guide has the exact JSON for each client, including the two that do not use the mcpServers key.

If a GUI client reports that the server failed to start, it is almost always PATH: a GUI-launched application does not inherit your shell's, so uvx may not be found. Use the absolute path (which uvx) as "command", and restart the application fully rather than reloading the window.

{
  "mcpServers": {
    "gopher": {
      "command": "uv",
      "args": ["--directory", "/path/to/gopher-mcp", "run", "task", "serve"]
    }
  }
}

Usage

The server registers eight MCP tools:

The four fetch tools are annotated read-only and open-world. The four certificate tools never touch the network, and each pair is split read from write so a client can gate the destructive one on its own.

Alongside them the server exposes one resource, gopher-mcp://policy, which renders the fetch policy this process is actually running with — the allowlists, caps and robots settings a refusal is decided from, with the two store paths reduced to <configured> / <default>. There is deliberately no tool that edits it: a fetched page talked into widening an allowlist would have widened it for every later fetch. Two prompts, Explore a capsule or Gopher hole and Summarize a gemlog or phlog, package the navigation and safety rules as a one-click starting point.

gopher_fetch Tool

Fetches Gopher menus, text files, or metadata by URL with comprehensive error handling and security safeguards.

Parameters:

• url (string, required): Full Gopher URL (e.g., gopher://gopher.floodgap.com/1/)
• search (string, optional): Terms for a type-7 (Index-Search) selector. They are percent-encoded and sent as the query string, so pass the user's words raw — a query holding #, +, & or non-ASCII is truncated or mangled when written into the URL by hand. RFC 1436 gives only type 7 a query field, so leave it unset for every other item type
• refresh (boolean, optional, default false): Skip the cached copy and re-fetch from the server
• offset (integer, optional, default 0): Continue a truncated result — pass the previous result's next_offset, which counts menu items for a menu

Response Types:

• MenuResult (kind: "menu"): For Gopher menus (type 1) and search results (type 7)
  • Structured menu items with type, title, selector, host and port, each with a next_url to follow. An empty next_url marks a display-only info line
• TextResult (kind: "text"): For text files (type 0)
  • Returns the text content with metadata
• BinaryResult (kind: "binary"): Metadata only for the binary item types (4, 5, 6, 9, g, I, d, s, ;, p, P, :, M, <)
  • Provides bytes and mime_type without downloading binary content
• ErrorResult (kind: "error"): For errors and unfetchable content
  • error.code and error.message; nothing was fetched. The interactive types (2 CSO, 8 telnet, T tn3270) have no fetchable body at all and answer NOT_FETCHABLE without opening a connection

gemini_fetch Tool

Fetches Gemini content with full TLS security, TOFU certificate validation, and native gemtext parsing.

Parameters:

• url (string, required): Full Gemini URL (e.g., gemini://geminiprotocol.net/)
• input (string, optional): Text to answer a Gemini input prompt (status 10/11); it is percent-encoded into the query string
• refresh (boolean, optional, default false): Skip the cached copy and re-fetch from the server
• offset (integer, optional, default 0): Continue a truncated result — pass the previous result's next_offset, which counts characters for a page body

Response Types: seven, one per kind.

• GeminiGemtextResult (kind: "gemtext"): For gemtext content (text/gemini)
  • Parsed document in document.lines and document.links, whose url fields are already resolved. A line carries its own type, content and whatever the marker cannot say (text, level, alt_text, language); there is no nested per-line object and no whole-document raw_content in the payload
• GeminiSuccessResult (kind: "success"): For other text content types
  • Decoded text in content, with MIME type information
• GeminiBinaryResult (kind: "binary"): For binary content
  • Metadata only — size and the detected mime_type, never the bytes. A 1 MB body would be ~350k tokens of base64 the model cannot render anyway
• GeminiInputResult (kind: "input"): For input requests (status 1x)
  • The capsule's prompt, with sensitive: true on status 11. Answer it by calling again with input=, never by hand-building a query string
• GeminiRedirectResult (kind: "redirect"): For redirects (status 3x, where 31 is permanent)
  • new_url is the target. Redirects are not followed for you, so the result also carries cross_host (the target belongs to a different party than the one you asked for) and scheme (anything but gemini leaves Geminispace and cannot be fetched with this tool). Follow at most five in a row and stop on a URL already seen
• GeminiErrorResult (kind: "error"): For errors (status 40-59), and for failures raised on this side of the wire — SSRF and allowlist refusals, a robots block, a certificate mismatch, a timeout
  • error.code and error.message, where message is written by this server. The capsule's own untrusted META text is kept apart in error.meta, so a hostile 51 <instruction> cannot be read as this server's guidance. Where a status has a defined remedy — the whole temporary 4x family included — that remedy is in error.next_step
• GeminiCertificateResult (kind: "certificate"): For certificate statuses (60-69)
  • Certificate requirement information, plus a next_step written by this server (message is the capsule's own text). A certificate that already exists for the host/port/path scope is attached automatically and the fetch path never creates one, so retrying unchanged returns status 60 again; gemini_client_cert_update mints one for that scope, but only once the user has agreed to hold a persistent identity on that capsule.

GeminiErrorResult is an alias for the same ErrorResult model gopher_fetch returns, not a separate type — its error object simply carries the extra status, temporary and meta keys when the capsule actually answered.

Gemini results name the content length size where the Gopher results name the same fact bytes. One concept, two wire names, kept apart only because renaming either would break every existing consumer.

Cached Results and refresh

Successful bodies are cached per protocol for a few minutes. A result that came from the cache says so, so a replay is never mistaken for the current state of a resource:

• cached — true when the result was replayed from the local cache
• cached_at — when that copy was actually fetched, as an ISO-8601 UTC timestamp (2026-09-02T12:00:00+00:00)
• cache_age_seconds — how old the copy was when it was returned

These appear only on the kinds that are actually cached (Gopher menu, text, binary; Gemini gemtext, success, binary). Errors, redirects and input/certificate prompts are never cached.

Pass refresh: true when the user wants the current state — it skips the cache for that one call and still stores the fresh response. All four fetch tools take it, the batch pair included.

Truncated Results and offset

Menus and page bodies are capped before they reach the model (*_MAX_RENDERED_CHARS, GOPHER_MAX_MENU_ITEMS), but a cap is not a dead end. A result cut short sets truncated: true and says where to resume:

• next_offset — where the part that was cut begins, or null when there is nothing more
• total_items (Gopher menu) / total_chars (text, success, gemtext) — how big the whole resource is. total_items is null when the directory was larger than the render cap, because the total is not counted in that case

Call the same tool again with offset set to the previous next_offset and keep going until next_offset comes back null. The unit is items for a menu and characters for a body; bytes and size are byte counts and are never offsets. For gemtext, a window ends on the last complete line, so consecutive windows abut exactly and half a link never parses as a whole one.

Neither batch tool takes offset: one offset cannot mean anything sensible across a list of different URLs. Continue a truncated batch item with the single-URL tool, which is where next_offset is answerable.

Gemini Trust-Store Tools

Gemini has no certificate authorities: the first certificate seen for a host is pinned, and every later connection must present the same one. When a host reissues its certificate — routine for self-signed certs, usually at expiry — the fetch fails with CERTIFICATE_CHANGED. Two tools handle that without hand-editing the trust store on disk — $XDG_DATA_HOME/gopher-mcp/tofu.json, falling back to ~/.local/share/gopher-mcp/, ~/Library/Application Support/gopher-mcp/ on macOS and %LOCALAPPDATA%\gopher-mcp\ on Windows, and overridable with GEMINI_TOFU_STORAGE_PATH. An install that already has ~/.gemini/tofu.json keeps using it exactly where it is, permanently: moving pins would lose them or, worse, make a pinned host look unpinned. The full rules are in Where Gemini state is stored.

• gemini_trust_list (read-only) reports what is pinned, optionally for one host: fingerprint, port, first/last seen and expiry as ISO-8601 UTC, plus a precomputed expired — an ended validity window makes a routine reissue the likely explanation for a changed fingerprint.
• gemini_trust_update (destructive) removes (action: "remove") or replaces (action: "pin") the pin of one named host. There is no wildcard.

A fingerprint change is also exactly what an active machine-in-the-middle attack looks like, and the two are indistinguishable from the client. So a pin is only ever changed after the user confirms the new certificate is expected — checked against the operator or another device, never on the say-so of a fetched page. To enforce that, action: "remove" requires the fingerprint currently pinned (as reported by gemini_trust_list); a mismatch returns FINGERPRINT_MISMATCH and changes nothing. On a client that supports MCP elicitation the change is also put to you before it is made, and declining returns USER_DECLINED with the pin untouched; a client without that capability is never asked, and behaves as it always has.

Gemini Client-Identity Tools

A client certificate is the other half of Gemini's certificate story, and the opposite direction: it is the identity this server presents to a capsule, not the one a capsule presents to us. Capsules with accounts ask for it with status 60 (certificate required). The fetch path attaches a certificate that already covers the requested scope but never creates one, so answering a 60 is an explicit call:

• gemini_client_cert_list (read-only) reports the scopes that hold an identity — each as a ready-to-use scope URL with its fingerprint, validity window and whether it has expired. Never a private key or its location.
• gemini_client_cert_update (destructive) creates the identity for the scope of a named gemini:// URL, or removes the one covering it. Like the trust-store tool, it asks first on a client that supports elicitation — creating an identity is a decision about being linkable across visits, and removing one destroys a private key that nothing can recreate.

The certificate covers that URL's path and everything below it, so gemini://host/app/page.gmi covers one page, gemini://host/app/ the section, and gemini://host/ the whole capsule. While it exists, every request in that scope carries it, which is what lets the capsule link those visits — so it is the user's decision, never a reaction to a page or META string asking for one. Creation refuses to replace a certificate already covering the scope, because the private key cannot be recovered, and action: "remove" requires the fingerprint being destroyed, exactly as the trust tools require the pinned one.

Example URLs to Try

Gopher Protocol

# Classic Gopher menu
gopher://gopher.floodgap.com/1/

# Gopher news and information
gopher://gopher.floodgap.com/1/gopher

# Search example (type 7)
gopher://gopher.floodgap.com/7/v2/vs

# Text file example
gopher://gopher.floodgap.com/0/gopher/welcome

Gemini Protocol

# Gemini protocol homepage
gemini://geminiprotocol.net/

# Gemini software directory
gemini://geminiprotocol.net/software/

# Example personal capsule
gemini://skyjake.fi/

# A large, browsable aggregator capsule
gemini://kennedy.gemi.dev/

Geminispace has no search engine this tool can drive. Kennedy and tlgs.one are worth browsing, but both Disallow: /search in their robots.txt, so a search URL on either comes back BLOCKED_BY_ROBOTS. That is a stop, not a setting to change: the operators asked automated clients to stay off those paths.

Example AI Interactions

Once configured, you can ask Claude:

Gopher Exploration:

• "Browse the main Gopher menu at gopher.floodgap.com"
• "Search for 'python' on the Veronica-2 search server"
• "Show me the welcome text from Floodgap's Gopher server"
• "What's available in the Gopher community directory?"

Gemini Exploration:

• "Fetch the Gemini protocol homepage"
• "Show me the software directory on geminiprotocol.net"
• "Browse the latest posts from a gemlog"
• "What's the difference between Gopher and Gemini protocols?"

Development

Task Runner

Every development command is a task in [tool.taskipy.tasks] (pyproject.toml), which is the single definition of each one — there is no second table to keep in sync. Run them the same way on every platform:

uv run task dev-setup     # install dependencies and pre-commit hooks
uv run task quality       # lint + typecheck + test
uv run task ci            # what CI runs: check + test-cov
uv run task help          # list every task (alias for `task --list`)

On Unix and macOS, make <command> is a thin catch-all onto the same table, and bare make runs help. uv run task help is the authoritative list; the tasks and the reasoning behind the ones that are not obvious are described in CONTRIBUTING.md.

Project Structure

gopher-mcp/
├── src/gopher_mcp/          # Main package
│   ├── __init__.py          # Package initialization
│   ├── __main__.py          # CLI entry point (--transport/--host/--port/--allowed-host)
│   ├── server.py            # FastMCP server + the eight MCP tool definitions
│   ├── client_base.py       # Shared fetch scaffolding for both clients
│   ├── gopher_client.py     # Gopher protocol client
│   ├── gopher_transport.py  # Low-level Gopher transport
│   ├── gopher_parse.py      # Gopher URL and menu parsing
│   ├── gemini_client.py     # Gemini protocol client
│   ├── gemini_tls.py        # Gemini TLS connection handling
│   ├── gemini_parse.py      # Gemini URL and response parsing
│   ├── gemtext.py           # Gemtext document parsing
│   ├── mime.py              # MIME type detection and filtering
│   ├── tofu.py              # Trust-on-First-Use certificate store
│   ├── client_certs.py      # Gemini client certificate storage
│   ├── identity.py          # Trust/identity decision and wording helpers
│   ├── ssrf.py              # SSRF protection / address filtering
│   ├── ratelimit.py         # Per-host rate limiting
│   ├── robots.py            # robots.txt fetching and policy gate
│   ├── cache.py             # Shared TTL + LRU response cache
│   ├── config.py            # Pydantic settings models
│   ├── models.py            # Pydantic data models
│   ├── helpers.py           # Shared URL/IO/sanitization helpers
│   └── utils.py             # Backward-compatible facade re-exporting the above
├── tests/                   # Comprehensive test suite
├── docs/                    # MkDocs documentation
├── scripts/                 # Development scripts
├── config/                  # Example configuration (example.env)
├── .github/workflows/       # CI/CD pipelines
├── Dockerfile               # Slim, non-root container image
├── Makefile                 # Unix/macOS shortcut onto the taskipy tasks
├── server.json              # MCP registry manifest
└── pyproject.toml           # Modern Python project config

Development Workflow

1. Setup: uv run task dev-setup - Install dependencies and pre-commit hooks
2. Code: Make your changes with full IDE support (type hints, linting)
3. Quality: uv run task quality - Run all quality checks (lint + typecheck + test)
4. Test: uv run task test-cov - Run tests with coverage reporting
5. Commit: Pre-commit hooks ensure code quality automatically

Testing

# Run all tests
uv run task test

# Run with coverage
uv run task test-cov

# Run specific test types
uv run task test-unit
uv run task test-integration

# Run one file
uv run pytest tests/test_server.py

Configuration

The server can be configured through environment variables for both protocols:

Gopher Configuration

Gemini Configuration

SSRF protection: by default both tools reject targets that resolve to loopback, link-local (including cloud metadata 169.254.169.254), or private/RFC1918 addresses. Set GOPHER_ALLOW_LOCAL_HOSTS / GEMINI_ALLOW_LOCAL_HOSTS to true only when you deliberately need to reach local hosts (e.g. testing a server on localhost).

Timeouts. *_TIMEOUT_SECONDS is one overall deadline per fetch, not a per-phase timeout. For Gemini, DNS, connect and TLS handshake, the trust-store write, send and read all draw down the same budget, and when robots checking is enabled the /robots.txt probe spends from it too — so a slow host cannot spend the full value on each step in turn.

List-valued variables. *_ALLOWED_HOSTS, *_ALLOWED_PORTS and GEMINI_DENIED_MIME_TYPES accept either the comma-separated form (a,b) or a JSON array (["a", "b"]); whitespace around entries is stripped. Leave one unset (or empty) to mean "no restriction". A value that is present but names no entries — " , ", or "$A,$B" where both shell variables are empty — is a startup error, because an empty allowlist cannot be told apart from an absent one and would silently drop the restriction you meant to apply. A port outside 1–65535 in an allowlist is a startup error for the same reason: it could never match, so every fetch would be refused at runtime instead.

Caching. *_CACHE_TTL_SECONDS=0 disables caching rather than storing entries that expire the instant they are written.

The tables above cover the most common settings. Additional options include robots policy caching and retry (*_ROBOTS_CACHE_TTL_SECONDS, *_ROBOTS_HONOR_AI_TOKENS, *_ROBOTS_FAILURE_BACKOFF_SECONDS), rendered-output limits (*_MAX_RENDERED_CHARS, GOPHER_MAX_MENU_ITEMS), Gemini TOFU/certificate storage paths and expiry policy (GEMINI_TOFU_STORAGE_PATH, GEMINI_TOFU_REJECT_EXPIRED, GEMINI_CLIENT_CERTS_STORAGE_PATH), MIME filtering (GEMINI_DENIED_MIME_TYPES), and server/logging settings under the GOPHER_MCP_ prefix. See the full Configuration Guide for every variable, its type, range, and default, or config/example.env for a ready-to-edit starting point.

Example Configuration

# Gopher settings
export GOPHER_MAX_RESPONSE_SIZE=2097152
export GOPHER_TIMEOUT_SECONDS=60
export GOPHER_CACHE_ENABLED=true
export GOPHER_ALLOWED_HOSTS="gopher.floodgap.com,gopher.quux.org"

# Gemini settings
export GEMINI_MAX_RESPONSE_SIZE=2097152
export GEMINI_TIMEOUT_SECONDS=60
export GEMINI_TOFU_ENABLED=true
export GEMINI_CLIENT_CERTS_ENABLED=true
export GEMINI_ALLOWED_HOSTS="geminiprotocol.net,skyjake.fi"

# Run with custom config
uv run task serve

Network Etiquette

Gopherspace and Geminispace are served largely by individuals running small machines. This server is built to be a guest there.

What this tool does and does not do

gopher-mcp fetches a resource when someone asks their assistant for it, and returns the content to that conversation. It does not train on what it fetches, archive it, rehost it, or make it searchable. It does not follow links on its own: every URL it retrieves was named by the caller.

Neither protocol has a user-agent field, so nothing identifies this client on the wire and a server cannot recognise or block it by name. That is a property of Gopher and Gemini, not a choice made here. What it does do, when robots checking is enabled below, is honour rules written against the token gopher-mcp, so an operator who wants to single it out has a way to.

Rate limiting

Both clients space out requests to the same host and cap how many fetches run at once. Unlike earlier versions, these are on by default: one request per second per host (*_REQUESTS_PER_MINUTE=60) and five concurrent fetches (*_MAX_CONCURRENT_REQUESTS=5). Set either to 0 to disable it. A Gemini server answering 44 SLOW_DOWN is always honoured regardless of these settings.

Robot exclusion (robots.txt)

The server fetches /robots.txt from the host root and honours it before retrieving anything. This is on by default: these are overwhelmingly hobbyist-run servers, and ignoring a stated policy is not a reasonable default for a tool an LLM drives unattended. Policies are cached per host for 24 hours (*_ROBOTS_CACHE_TTL_SECONDS), so the extra round-trip is paid once per host, not per fetch.

GOPHER_RESPECT_ROBOTS_TXT=false / GEMINI_RESPECT_ROBOTS_TXT=false turns it off, but the override is for a host you operate — a blanket Disallow: / on someone else's server is a decision, not a misconfiguration. The error messages are written that way too, because they are read by the model, not by you: a blocked fetch returns BLOCKED_BY_ROBOTS and says to stop and tell the user rather than retry or try another spelling of the path. A fetch refused because the policy could not be read at all returns ROBOTS_UNAVAILABLE instead — a separate code because it is transient and means nothing disallowed you.

Which convention applies depends on the protocol:

• Gemini follows the official companion specification, which defines the virtual agents archiver, indexer, researcher and webproxy. This server matches gopher-mcp, webproxy, indexer and *. It does not claim archiver or researcher: nothing here retains content, and researcher is defined for tools that operate without surfacing what they fetch.
• Gopher follows the convention Veronica-2 documents at gopher://gopher.floodgap.com/0/v2/help/indexer (written out rather than linked, because a gopher:// href does not survive PyPI's sanitizer). This server matches gopher-mcp and *. It does not claim veronica, which belongs to Floodgap's indexer.

Both use the original 1994 robots.txt grammar rather than RFC 9309, so only #, User-agent: and Disallow: are recognised and every other field, including Allow:, is ignored. This matters: an RFC 9309 parser would act on Allow: lines that authors on these networks expect to be dropped, making it more permissive than intended.

By default the server also honours rules naming AI crawler tokens such as ClaudeBot, GPTBot and CCBot (*_ROBOTS_HONOR_AI_TOKENS). These are not part of either protocol's convention, but an operator who wrote one meant "no LLM tooling", and that is the request being made.

When a policy cannot be retrieved at all, the host is left alone for a short while before being probed again (*_ROBOTS_FAILURE_BACKOFF_SECONDS, 60s by default) rather than paying a fresh connect timeout on every request. Set it to 0 to re-probe immediately, or raise it if you routinely fetch from hosts that are down. A Gemini capsule answering 44 SLOW_DOWN is the exception: it named its own retry period, so that is used instead.

Two known limitations, both documented rather than papered over:

• Gopher fails open. The protocol has no status codes, so a missing selector, an error document and an empty file are indistinguishable on the wire. RFC 9309 §2.3.1.4 would have an unreachable policy deny everything, which would block most of Gopherspace. Instead the parser is lenient: content that yields no User-agent: group imposes no rules. Gemini, which does have status codes, fails closed and treats 51 NOT FOUND as "no policy". Note that failing closed covers more than a 4x status: a capsule that is unreachable — connection refused, TLS failure, timeout, malformed reply — also has no retrievable policy, so it is refused too — but under the separate ROBOTS_UNAVAILABLE code, since the capsule never disallowed anything. The message names the underlying cause, and turning robots checking off will not make such a capsule reachable.
• Gopher path rules are best-effort. A Gopher URI carries the item type as the first path character, so gopher://host/1/archive has the URI path /1/archive but the on-wire selector /archive. Rules are tested against both spellings, but Disallow: / is the only form guaranteed to behave the way its author expects.

There is no per-directory or per-user robots.txt. Neither protocol convention nor RFC 9309 §2.3 defines one; on shared hosts the established pattern is a single file at the root using path prefixes.

Contributing

We welcome contributions! Please see our Contributing Guidelines for details.

Quick Contribution Steps

1. Fork the repository on GitHub
2. Clone your fork: git clone https://github.com/your-username/gopher-mcp.git
3. Setup development environment: uv run task dev-setup
4. Create a feature branch: git checkout -b feature/amazing-feature
5. Make your changes with tests
6. Quality check: uv run task quality
7. Commit your changes: git commit -m 'Add amazing feature'
8. Push to your fork: git push origin feature/amazing-feature
9. Submit a pull request with a clear description

Development Standards

• Type hints for all functions and methods
• Comprehensive tests (CI enforces a minimum of 95% coverage)
• Documentation for all public APIs
• Security considerations for all network operations
• Cross-platform compatibility (Windows, macOS, Linux)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Model Context Protocol by Anthropic - The foundation that makes this integration possible
• MCP Python SDK - Its bundled FastMCP (mcp.server.fastmcp) is what this server is built on, pinned to mcp>=1.28.1,<2
• The Gopher Protocol Community - Keeping the spirit of the early internet alive

Related Projects

• Model Context Protocol Servers - Official MCP server implementations
• Awesome MCP Servers - Curated list of MCP servers
• Claude Desktop - AI assistant that supports MCP

Support

• Bug Reports: GitHub Issues
• Feature Requests: Open a feature request
• Questions: Ask a question
• Documentation: Project Docs
• Community: MCP Discord

Whichever you open, include the version: gopher-mcp --version reports it and works for the uvx and Docker installs, where there is no checkout to import the package from.