Agent OS: the agent gets smarter on its own. We just hold the line: Interview-gated, staged evaluation, budgeted evolution loop. MCP server, 14 runtimes: Claude Code, Codex CLI, Gemini CLI, OpenCode, Copilot, Kiro and more.
- vulnerabilities attested + 18
- capability exposure inferred + 35
- recent drift inferred + 20
- tool safety inferred + 12
- trust mitigators mixed − 16
attested 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 →
- D · 77 → D · 69
- D · 69 → D · 77
- C · 57 → D · 69
- D · 77 → C · 57
- D · 69 → D · 77
- C · 45 → D · 69
- C · 50 → C · 45
- D · 62 → C · 50
- D · 74 → D · 62
- high dangerous code
committed secret: Slack token, GitLab token, Google API key, OpenAI key · dynamic exec: unsafe yaml.load(), __import__ sink, __import__() · credential logged in 1 file(s)
analyzed commit 7aa307c · analyzer v33 · 16h ago
skills & prompt files 26
- agent-rules Q00-ouroboros-7aa307c/AGENTS.md
- agent-rules Q00-ouroboros-7aa307c/CLAUDE.md
- agent-rules Q00-ouroboros-7aa307c/docs/runtime-guides/gemini.md
- skill Q00-ouroboros-7aa307c/skills/auto/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/brownfield/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/cancel/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/config/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/evaluate/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/evolve/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/help/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/interview/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/ooo/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/pm/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/publish/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/qa/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/ralph/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/resume-session/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/run/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/seed/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/setup/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/status/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/tutorial/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/unstuck/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/update/SKILL.md
- skill Q00-ouroboros-7aa307c/skills/welcome/SKILL.md
- skill Q00-ouroboros-7aa307c/tests/fixtures/router/skills/frontmatter-body/run/SKILL.md
danger signals13
- dynamic code execution unsafe yaml.load() Q00-ouroboros-7aa307c/src/ouroboros/cli/runtime_activation.py :343
loaded = yaml.load(snapshot.contents.decode("utf-8"), Loader=_UniqueKeyLoader) - dynamic code execution unsafe yaml.load() Q00-ouroboros-7aa307c/tests/unit/integrations/test_dsh_plugin_bundle.py :55
return list(yaml.load(text, Loader=_JsTolerantLoader)) - dynamic code execution __import__ sink Q00-ouroboros-7aa307c/tests/unit/orchestrator/test_verifier.py :314
__import__("subprocess").TimeoutExpired("pytest", 30), - dynamic code execution __import__() Q00-ouroboros-7aa307c/tests/unit/router/test_dispatch.py :356
wraps=__import__( - suspicious endpoint us.i.posthog.com (telemetry) Q00-ouroboros-7aa307c/src/ouroboros/telemetry.py :49
_DEFAULT_HOST = "https://us.i.posthog.com" - suspicious endpoint 169.254.169.254 (cloud metadata) Q00-ouroboros-7aa307c/tests/unit/mcp/test_types.py :195
url="http://169.254.169.254/latest/meta-data/", - suspicious endpoint 8.8.8.8 Q00-ouroboros-7aa307c/tests/unit/mcp/test_types.py :268
url="https://8.8.8.8/mcp", - committed secret Slack token Q00-ouroboros-7aa307c/tests/unit/auto/test_state_provenance.py :35
xoxb-s…(16 chars, redacted) - committed secret GitLab token Q00-ouroboros-7aa307c/tests/unit/core/test_security.py :142
glpat-…(32 chars, redacted) - committed secret Google API key Q00-ouroboros-7aa307c/tests/unit/mcp/resources/test_handlers.py :460
AIzaAA…(39 chars, redacted) - committed secret GitLab token Q00-ouroboros-7aa307c/tests/unit/orchestrator/test_route_policy.py :438
glpat-…(32 chars, redacted) - committed secret OpenAI key Q00-ouroboros-7aa307c/tests/unit/plugin/test_firewall_tool_call_dispatch.py :474
sk-thi…(37 chars, redacted) - credential in logs credential in log Q00-ouroboros-7aa307c/tests/unit/observability/test_logging.py :1247
log.info("config.loaded", api_key="sk-1234567890abcdef")
- recent drift +20 capability drift →
[](https://mcpobservatory.com/servers/github:Q00/ouroboros/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 Q00.