Local-first, safety-gated CLI and MCP harness for bounded DeepSeek batch and corpus workloads.
- capability exposure inferred + 35
- recent drift inferred + 12
- tool safety inferred + 12
- trust mitigators mixed − 3
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.
graded 4m ago · see ecosystem CVEs →
- B · 32 → C · 56
- B · 25 → B · 32
- B · 28 → B · 25
No known CVEs for this server.
- high dangerous code
committed secret: OpenAI key
analyzed commit 5a57549 · analyzer v28 · 3h ago
danger signals1
- committed secret OpenAI key tylerbuilds-deepseek-harness-5a57549/test/adversarial/test_adversarial_transport.ts :501
sk-pro…(30 chars, redacted)
- recent drift +12 capability drift →
[](https://mcpobservatory.com/servers/github:tylerbuilds/deepseek-harness/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 tylerbuilds.