Skip to content

Releases

renderfetch-mcp 2.0.0

Choose a tag to compare

@lalbers lalbers released this 28 Oct 21:16
v2.0.0
1d2b429

A security hardening release with breaking configuration and output changes.
Read Upgrading from 1.0.0 before rolling it out.

Security

  • Connection-level SSRF protection: every render gets a loopback-only egress proxy
    that validates the complete DNS answer, connects to the checked numeric IP and
    keeps Host/TLS validation on the original name. This closes DNS rebinding and
    covers redirects and subresources that Playwright routing never sees.
  • Service workers, WebSockets, downloads, extra pages, QUIC and non-proxied
    WebRTC UDP are blocked, and so are write-method requests.
  • Untrusted-output pipeline: consistent sanitizing across formats, Unicode
    normalization before matching, screening of the full bounded document before
    truncation, and page titles/URLs only inside the untrusted envelope. Errors no
    longer reflect page content.
  • OAuth: a required mcp:fetch scope, sessions bound to client and identity,
    client-bound revocation, refresh-token families with replay detection and a
    fixed absolute expiry, and restart-safe one-use consent.
  • The consent login limit counts only failed same-origin credential attempts, so
    anonymous junk requests can no longer lock the owner out.
  • Dynamic client registrations that never obtain a token are pruned after
    OAUTH_UNUSED_CLIENT_TTL, so /register cannot be filled up permanently.
  • Secret comparisons no longer leak the secret's length.
  • Container deployments run Chromium's renderers in its sandbox, via
    deploy/seccomp-chromium.json. The container smoke test verifies it.
  • Dependencies updated, npm audit clean: MCP SDK 1.32.2, Playwright 1.64.1,
    zod 4, jose 6, jsdom 30, express 5.3.

Added

  • cookie_banner tool option (hide by default) removes recognized consent
    overlays from the output snapshot without clicking or changing consent.
  • Settings: ALLOWED_ORIGINS, TRUST_PROXY, FETCH_ALLOWED_PORTS,
    FETCH_TIMEOUT_MS, FETCH_MAX_REQUESTS, FETCH_MAX_HTML_BYTES,
    FETCH_MAX_TRANSFER_BYTES, BROWSER_MAX_QUEUE, MCP_SESSION_TTL_MS,
    MCP_MAX_SESSIONS, MCP_MAX_SESSIONS_PER_CLIENT, OAUTH_MAX_CLIENTS,
    OAUTH_UNUSED_CLIENT_TTL and CLIENT_SECRET_TTL.
  • Real-browser security regression tests run in CI.

Changed

  • Page text is returned in both content and structuredContent.text. Some
    clients, including the claude.ai connector, render only structuredContent.
  • Client secrets issued by dynamic registration live for one year
    (CLIENT_SECRET_TTL) instead of the SDK's 30 days.
  • Client registrations are logged by client_id only, without client-supplied
    names or redirect URIs.
  • Invalid configuration (short secrets, malformed booleans, out-of-range numbers)
    fails at startup. An empty STATIC_BEARER_TOKEN disables static bearer auth.

Removed

  • structuredContent.title and structuredContent.final_url. Read them from the
    fenced text instead.

Upgrading from 1.0.0

  • Node 22.22.2+, 24.15+ or 26+ is required (Node 24 recommended). Use
    the committed lockfile.
  • Back up the OAuth database first: the refresh-token migration is one-way.
    Rotations made under 1.0.0 cannot be recognized as replays afterwards.
  • Existing tokens without mcp:fetch must reconnect. Refresh requires
    offline_access.
  • Configuration is validated at startup. The example values from 1.0.0's
    .env.example (change-me…) are rejected. AUTH_PASSWORD needs 12+
    characters. STATIC_BEARER_TOKEN needs 32+ printable ASCII characters without
    whitespace. PUBLIC_BASE_URL must be an HTTPS origin (HTTP only for localhost).
    Booleans must be true/false (or 1/0, yes/no, on/off), and numeric
    settings have bounds; the startup error names any value out of range.
  • Clients registered under 1.0.0 keep their 30-day client secret. A connector may
    need to be re-linked once when it expires; new registrations get one year.
  • structuredContent.title and .final_url are gone; read the fenced text.
  • Behind a reverse proxy, set TRUST_PROXY to its address. Otherwise all clients
    share one rate-limit bucket and anyone can lock out the login.
  • Browser tools such as MCP Inspector need an explicit ALLOWED_ORIGINS entry.
  • Nonstandard destination ports require FETCH_ALLOWED_PORTS. Unsafe schemes and
    URL credentials are rejected even with private access enabled.
  • JavaScript still runs, but service workers, WebSockets and write-method requests
    are blocked. Some POST-backed applications render incompletely.
  • The code default of CHROMIUM_NO_SANDBOX changed from true to false. Add
    deploy/seccomp-chromium.json to your container setup (see the README).
    Without it Chromium cannot start sandboxed: /healthz still reports ok, but
    every fetch fails. 1.0.0's example .env set CHROMIUM_NO_SANDBOX=true
    explicitly: change it to false after adding the profile, or Chromium keeps
    running unsandboxed. Use true only as a last resort.
  • Cookie cleanup defaults to hide. Use cookie_banner: "off" for unmodified
    consent text.

Full changelog: v1.0.0...v2.0.0

renderfetch-mcp 1.0.0

Choose a tag to compare

@lalbers lalbers released this 28 Oct 21:16
v1.0.0
0a8b354

First public release: a headless-Chromium fetch_url MCP tool with readability
extraction to Markdown, a self-contained OAuth 2.1 server (dynamic client
registration, PKCE, refresh-token rotation, SQLite store), prompt-injection
screening, an SSRF guard and rootless-Podman deployment examples.