Open-source AI-layer security testing — probes your AI agent for weaknesses and writes a validated regression test for each one it finds, gating CI.
- capability exposure inferred + 22
- tool safety inferred + 12
- trust mitigators mixed − 9
inferred mixed
The A–E grade is our heuristic synthesis — a "review this" prompt, not a verdict. Each factor is tagged by what backs it: attested (a verifiable record), reported (a third party's claim), or inferred (our own heuristic, e.g. permissions). See methodology.
grade last moved 3d ago · see ecosystem CVEs →
- B · 31 → B · 25
No known CVEs for this server.
- high dangerous code
committed secret: Anthropic key
analyzed commit 4861414 · analyzer v33 · 4d ago
danger signals3
- suspicious endpoint 169.254.169.254 (cloud metadata) Abidemialade-mylonite-4861414/tests/scan/test_control_shim.py :545
assert not host_allowed("http://169.254.169.254/latest/meta-data/", al) - committed secret Anthropic key Abidemialade-mylonite-4861414/tests/scan/test_engine.py :447
sk-ant…(36 chars, redacted) - committed secret Anthropic key Abidemialade-mylonite-4861414/tests/test_cli_keyless.py :428
sk-ant…(37 chars, redacted)
[](https://mcpobservatory.com/servers/github:Abidemialade/mylonite/security) Heuristic, inferred signals — false positives (legitimately powerful tools, forks, language ports) are expected. Treat each as "review this", not a verdict. See the ecosystem-wide picture on the security hub, or the fleet security of Abidemialade.