Unreleased: not public yet. Maintainers have not reviewed this result.
Results / fetch-mcp
fetch-mcp
A
Protocol4/4 points
F
Network1.5/4 pointscritical: net-loopback-ssrf
Coverage: covered (2 of 2 tools reached with a valid call).
Protocol checks
| Check | Verdict | What happened |
|---|---|---|
| protocol Valid MCP handshake and tool schemas | pass | handshake ok, 2 well-formed tool(s) |
| tool-poisoningcritical No tool-poisoning indicators in tool metadata | pass | scanned 2 tool definition(s), including parameter text: nothing suspicious |
| robust-input Survives malformed tool input | pass | 8 invalid call(s) rejected cleanly |
| secret-echocritical No environment secret echoed in responses | pass | canary secrets absent from tools/list and 8 captured response(s) |
Network checks
| Check | Verdict | What happened |
|---|---|---|
| net-loopback-ssrfcritical Refuses loopback targets (SSRF) | fail | fetch_url, fetch_youtube_transcript contacted a loopback listener the harness owns (loopback, localhost, ipv6, short, hex, decimal, unspecified, redirect): 16 of 18 probes reached it |
| net-scheme-allowlistcritical Only http(s) schemes accepted | pass | 14 probes with file:, ftp:, gopher:, dict:, data: and javascript: URLs had no observable effect (a scheme the HTTP library does not support passes trivially) |
| net-caps Response size and time are capped | fail | size: the server read the whole 64 MiB body (the listener sent 64 of 64 MiB) and returned 0.00 MiB: it capped its reply, not what it reads; time: no reply within 18 s (timeout_probe_sec 10 + 8 s) |
| net-redirect Redirects are re-validated | partial | redirects are followed; whether the target is re-validated could not be told apart, because the server accepted every loopback spelling directly |
What was not covered
- Database: Not applicable. Server config has no {dsn}/{host}/... Placeholder, so the harness cannot start the server once per role
- HTTP: Not applicable. The server is started over stdio; the http pack needs server.url
- Files: Not applicable. No tool is classified as taking a file path with medium confidence or higher
- Git: Not applicable. No tool is classified as taking a git repository path with medium confidence or higher
- SQLite: Not applicable. No tool takes a SQL string
- Effects: Not applicable. No tool declares annotations, so there is nothing to compare with what the sandbox observed
- Tools that a server lists only for clients that declare roots, sampling or elicitation capabilities are not seen: the harness declares none
- SSRF: link-local and private-range addresses are never contacted; redirects from a public host to loopback and DNS rebinding are not tested (the harness contacts only its own loopback listener)
- The sandbox is a convention the harness follows, not a security boundary. A malicious server runs with your user's privileges, sees the real file system and network, and can ignore cwd, HOME and PATH. Run untrusted servers in a throwaway VM or container.
- Prompt injection carried in tool output (only tool metadata is scanned)
- Authentication and transport security
- Audit logging (not measured by these packs; only the db pack has an advisory audit check)
Run it yourself
Started with the standard settings like this; put it in sealmark.yaml as the server:
npx -y fetch-mcp@0.0.5
Each of 613 verdicts was checked independently, and Sealmark was wrong on some: how reliable Sealmark is.
Think a verdict is wrong, or want a server removed? Email disclosure@sealmark.dev with the server version. We recheck, correct the page if we were wrong, and add your reply. See the disclosure policy.