Kiln-AI/Kiln
Build, Evaluate, and Optimize AI Systems. Includes evals, RAG, agents, fine-tuning, synthetic data generation, dataset management, MCP, and more.
- capability exposure inferred + 35
- recent drift inferred + 20
- tool safety inferred + 12
- trust mitigators mixed − 8
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 →
- C · 51 → C · 59
- A · 0 → C · 51
No known CVEs for this server.
- high dangerous code
dynamic exec: eval()/exec(), pickle.loads()
analyzed commit 9893493 · analyzer v33 · 2h ago
skills & prompt files 11
- skill Kiln-AI-Kiln-9893493/.agents/playwright_project/skills/197586632641 - escalation-and-reply-tone/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/playwright_project/skills/255490964102 - ticket-routing-playbook/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/claude-maintain-models/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/kiln-check-deprecation/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/kiln-check-finetune-deprecation/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/kiln-prerelease-check/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/open-pr/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/playwright/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/qa/SKILL.md
- skill Kiln-AI-Kiln-9893493/.agents/skills/release-digest/SKILL.md
- agent-rules Kiln-AI-Kiln-9893493/AGENTS.md
danger signals5
- dynamic code execution eval()/exec() Kiln-AI-Kiln-9893493/libs/core/kiln_ai/adapters/eval/sandbox_worker.py :57
exec(compile(code, "<code_eval>", "exec"), namespace) - dynamic code execution pickle.loads() Kiln-AI-Kiln-9893493/libs/core/kiln_ai/adapters/eval/test_g_eval.py :248
run_output = pickle.loads(serialized_run_output) - dynamic code execution pickle.loads() Kiln-AI-Kiln-9893493/libs/core/kiln_ai/adapters/eval/test_g_eval_characterization.py :166
run_output = pickle.loads(serialized_run_output) - dynamic code execution pickle.loads() Kiln-AI-Kiln-9893493/libs/core/kiln_ai/datamodel/test_usage.py :64
restored = pickle.loads(legacy_bytes) - dynamic code execution eval()/exec() Kiln-AI-Kiln-9893493/libs/core/kiln_ai/sandbox/worker.py :46
exec(compile(code, "<code_tool>", "exec"), namespace)
- recent drift +20 capability drift →
[](https://mcpobservatory.com/servers/github:Kiln-AI/Kiln/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 Kiln-AI.