cameronrye/gopher-mcp

v0.4.2 Security

This release includes 4 security fixes for security teams reviewing exposed deployments.

Published 1mo MCP Browser & Automation

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

This release patches 4 known CVEs

Topics

gemini gemini-protocol gopher gopher-protocol mcp mcp-server

Affected surfaces

crypto_tls breaking_upgrade

ReleasePort's take

Light signal

editorial:auto 1mo

The release makes Gemini TLS connects cancellable via native asyncio and stops leaking sensitive query data (status‑11) from logs, caches, and model inputs.

Why it matters: Enabling cancellable TLS connections improves deadline handling; preventing SENSITIVE_INPUT leakage safeguards user privacy in logs and caches.

Summary

AI summary

Updates Version, Commit, and Workflow across a mixed release.

Changes in this release

Type	Severity	Summary	CVE
Security
Security	Critical	Move Gemini TLS transport to native asyncio, making connects cancellable at request deadline. Move Gemini TLS transport to native asyncio, making connects cancellable at request deadline. Source: llm_adapter@2026-06-09 Confidence: high	—
Security	High	Stop leaking Gemini status‑11 (`SENSITIVE_INPUT`) query data in logs, model input, and cache keys. Stop leaking Gemini status‑11 (`SENSITIVE_INPUT`) query data in logs, model input, and cache keys. Source: llm_adapter@2026-06-09 Confidence: high	—
Security	High	Reject not‑yet‑valid certificates on TOFU first use by default, previously only warned unless `reject_expired` was set. Reject not‑yet‑valid certificates on TOFU first use by default, previously only warned unless `reject_expired` was set. Source: llm_adapter@2026-06-09 Confidence: high	—
Security	Medium	Reject empty and self‑referential Gemini redirects to prevent unbounded client re‑fetch loops. Reject empty and self‑referential Gemini redirects to prevent unbounded client re‑fetch loops. Source: llm_adapter@2026-06-09 Confidence: high	—
Feature	Medium	Add `GOPHER_MAX_MENU_ITEMS` config (default 1000) to cap returned Gopher menu items, mirroring gemtext character caps. Add `GOPHER_MAX_MENU_ITEMS` config (default 1000) to cap returned Gopher menu items, mirroring gemtext character caps. Source: llm_adapter@2026-06-09 Confidence: high	—
Feature	Medium	Add optional positive port allowlist (`GOPHER_ALLOWED_PORTS` / `GEMINI_ALLOWED_PORTS`) to close arbitrary‑port scanning gap. Add optional positive port allowlist (`GOPHER_ALLOWED_PORTS` / `GEMINI_ALLOWED_PORTS`) to close arbitrary‑port scanning gap. Source: llm_adapter@2026-06-09 Confidence: high	—
Dependency	Low	Drop hardcoded TLS 1.2 cipher allow‑list, using Python's secure defaults for better Gemini server interoperability. Drop hardcoded TLS 1.2 cipher allow‑list, using Python's secure defaults for better Gemini server interoperability. Source: llm_adapter@2026-06-09 Confidence: low	—
Deprecation	Medium	Remove unused `http`/`aiohttp` optional dependency, never‑read `development_mode` setting, and test‑only `create_tls_client` factory. Remove unused `http`/`aiohttp` optional dependency, never‑read `development_mode` setting, and test‑only `create_tls_client` factory. Source: llm_adapter@2026-06-09 Confidence: high	—
Deprecation	Medium	Remove `GeminiGemtextResult` helper properties `.summary`, `.plain_text`, and `.structured_content` as they were never serialized. Remove `GeminiGemtextResult` helper properties `.summary`, `.plain_text`, and `.structured_content` as they were never serialized. Source: llm_adapter@2026-06-09 Confidence: high	—
Bugfix	Medium	Apply `max_rendered_chars` cap to `text/gemini` responses, adding a `truncated` flag for both gemtext and menu results. Apply `max_rendered_chars` cap to `text/gemini` responses, adding a `truncated` flag for both gemtext and menu results. Source: llm_adapter@2026-06-09 Confidence: high	—

Full changelog

Security

Move the Gemini TLS transport to native asyncio (asyncio.open_connection
with ssl=), so connect, handshake and every read are genuinely cancellable:
a slow-loris or stalled peer is now cut off at the request deadline instead of
parking a worker thread on a blocking recv. The previous design ran blocking
socket I/O on a thread pool shared with DNS resolution (the SSRF guard for both
protocols), so a handful of slow Gemini reads could stall DNS for every request
and escalate one slow server into a whole-server denial of service.
Stop leaking a Gemini status-11 (SENSITIVE_INPUT) answer: the percent-encoded
query is no longer written to logs, reflected back to the model via
requestInfo, or retained in a cache key.
Reject not-yet-valid certificates on TOFU first use by default (previously
pinned with only a warning unless reject_expired was set).
Reject empty and self-referential Gemini redirects (INVALID_REDIRECT) so a
malformed 3x response cannot drive an unbounded client re-fetch loop.

Added

GOPHER_MAX_MENU_ITEMS (default 1000): caps the number of Gopher menu items
returned to the model, mirroring the existing text/gemtext character cap.
Optional positive port allowlist (GOPHER_ALLOWED_PORTS /
GEMINI_ALLOWED_PORTS) to close the arbitrary-port port-scanning gap left by
the dangerous-ports denylist.

Changed

Apply the max_rendered_chars cap to text/gemini responses (previously only
text/*), so a large gemtext page no longer floods the model context; both
gemtext and menu results now carry a truncated flag.
Drop the over-strict hardcoded TLS 1.2 cipher allow-list in favour of Python's
secure defaults, improving interop with conforming Gemini servers.
De-duplicate the two batch-fetch tools (gopher_batch_fetch /
gemini_batch_fetch) onto a single shared implementation so the batch
error/contract behaviour has one source of truth (no behaviour change).

Removed

The unused http/aiohttp optional dependency, the never-read
development_mode setting, and the test-only create_tls_client factory.
The GeminiGemtextResult.summary, .plain_text and .structured_content
helper properties: they were never serialized by model_dump() so the MCP
tools never exposed them (dead LLM-facing API). The parsed document and
raw_content carry the same information.

Fixed

Correct the LLM-facing server instructions, gopher_fetch parameter
description, and AI Assistant Guide to reference the real serialized
next_url menu field (not the nextUrl/url names that never appear in the
output).
Sync server.json to the released version and replace placeholder

Release Information:

Version: 0.4.2
Release Date: 2026-06-09
Commit: 93be258
Workflow: 27209467836

Installation:

pip install gopher-mcp==0.4.2

Verification:
All packages are signed and can be verified using Sigstore.

Breaking Changes

Removed `http`/`aiohttp` optional dependency, `development_mode` setting, and test‑only `create_tls_client` factory.
Removed unused `GeminiGemtextResult.summary`, `.plain_text`, and `.structured_content` properties.

Security Fixes

Gemini TLS transport now uses native asyncio for cancellable I/O, preventing worker‑thread stalls and denial‑of‑service from slow peers.
Stopped leaking Gemini status‑11 (`SENSITIVE_INPUT`) data in logs, `requestInfo`, and cache keys.
Reject not‑yet‑valid certificates on TOFU first use by default; previously only warned unless `reject_expired` was set.
Reject empty and self‑referential Gemini redirects to prevent unbounded client re-fetch loops.

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track cameronrye/gopher-mcp

Get notified when new releases ship.

About cameronrye/gopher-mcp

Modern, cross-platform MCP server enabling AI assistants to browse and interact with both Gopher protocol and Gemini protocol resources safely and efficiently. Features dual protocol support, TLS security, and structured content extraction.

All releases →

Related context

Related tools

Earlier breaking changes

v0.5.0 Changes Gemini binary success responses to return only metadata, dropping base64‑inlined bodies.
v0.3.0 Reject `--mount-path` for transports that ignore it instead of silently dropping.
v0.3.0 Server settings now read under `GOPHER_MCP_` environment prefix; update any existing env vars accordingly.