Manifest-driven 21-stage agent pipeline for Python — 5 LLM backends (Anthropic / OpenAI / Google / vLLM / Claude Code CLI), tools, skills, memory, sandboxed CLI runs, MCP. The engine behind Geny. Apache-2.0.
- capability exposure inferred + 35
- recent drift inferred + 20
- tool safety inferred + 5
- 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 42s ago · see ecosystem CVEs →
- C · 49 → C · 57
No known CVEs for this server.
- medium dangerous code
dynamic exec: eval()/exec()
analyzed commit 0dcfbcf · analyzer v28 · 2d ago
skills & prompt files 10
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/batch/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/debug/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/environment/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/loop/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/lorem-ipsum/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/simplify/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/skillify/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/stuck/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/tool-builder/SKILL.md
- skill CocoRoF-geny-executor-0dcfbcf/src/geny_executor/skills/bundled/verify/SKILL.md
danger signals8
- dynamic code execution eval()/exec() CocoRoF-geny-executor-0dcfbcf/src/geny_executor/tools/adhoc.py :401
exec(compile(config.code, "<adhoc>", "exec"), restricted_globals) - dynamic code execution eval()/exec() CocoRoF-geny-executor-0dcfbcf/tests/unit/test_phase2_agent_loop.py :64
result = eval(input["expression"]) # noqa: S307 - dynamic code execution eval()/exec() CocoRoF-geny-executor-0dcfbcf/tests/unit/test_phase6_integration.py :64
result = eval(input["expression"]) # Safe in test context - suspicious endpoint api.telegram.org CocoRoF-geny-executor-0dcfbcf/src/geny_executor/channels/built_in.py :159
f"https://api.telegram.org/bot{self._token}/sendMessage", json=body - suspicious endpoint api.telegram.org CocoRoF-geny-executor-0dcfbcf/src/geny_executor/gateway/telegram.py :23
_API = "https://api.telegram.org" - suspicious endpoint api.telegram.org CocoRoF-geny-executor-0dcfbcf/tests/unit/test_builtin_channels.py :59
assert calls["url"] == "https://api.telegram.org/bot123:abc/sendMessage" - suspicious endpoint 169.254.169.254 (cloud metadata) CocoRoF-geny-executor-0dcfbcf/tests/unit/test_phase7_security.py :72
validate_url("http://169.254.169.254/latest/meta-data/") - suspicious endpoint 169.254.169.254 (cloud metadata) CocoRoF-geny-executor-0dcfbcf/tests/unit/test_security_hardening.py :14
"http://169.254.169.254/latest/meta-data/", # cloud metadata
- recent drift +20 capability drift →
[](https://mcpobservatory.com/servers/github:CocoRoF/geny-executor/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 CocoRoF.