semcod/koru

github

github Python analyzed 628d1ff deep scan · partial (1,158/3,958 files) unconfirmed MCP

Closed-loop automation across semcod/ repositories.

maintainer: semcod
licence: Apache-2.0
first seen: 2026-06-04
last seen: 2026-10-15
releases · 30d: 0
github stars: 2 reported
short id

risk

64/100 · heuristic grade

D high inferred analyzer v33 (current)

capability exposure inferred + 35
recent drift inferred + 20
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.

risk trajectory

4 movements

14w ago C · 57 → D · 64
16w ago D · 64 → C · 57
17w ago C · 56 → D · 64
17w ago A · 0 → C · 56

capability exposure

grade factor +35

Inferred surface — each links to servers holding it:

vulnerabilities

0 CVEs

No known CVEs for this server.

tool safety

2 findings · grade factor +12

high dangerous code

dynamic exec: __import__(), __import__ sink, unsafe yaml.load(), eval()/exec()
medium dangerous code

env-secret-flows-to-network-py: An environment value reaches a network call. Review whether it is a credential leaving the process; the ordinary API-wrapper shape (read a key, call

skills & danger signals

github-tarball

prompt-surface shipped agent-instruction files + hidden-content / dangerous-code findings — quoted from the analyzed source

analyzed commit 628d1ff · analyzer v33 · 1d ago

skills & prompt files 5

agent-rules .cursor/rules/new-project-standard.mdc
agent-rules .github/copilot-instructions.md
agent-rules AGENTS.md
agent-rules CLAUDE.md
agent-rules GEMINI.md

danger signals11

dynamic code execution __import__() packages/coru/src/coru/cli_startup.py :486 __import__(module_name)
dynamic code execution __import__ sink src/koru/autopilot/diagnose_vdisplay_cli.py :47 source = __import__("os").environ.get("KORU_VDISPLAY_SOURCE", "DP-1")
dynamic code execution __import__ sink src/koru/autopilot/vdisplay_up_cli.py :12 explicit = __import__("os").environ.get("KORU_VDISPLAY_SOURCE", "").strip()
dynamic code execution __import__() src/koru/cli.py :150 return getattr(__import__(module_name, fromlist=[attr_name]), attr_name)(argv)
dynamic code execution __import__() src/koru/cli/__init__.py :103 value = getattr(__import__(module_name, fromlist=[attr_name]), attr_name)
dynamic code execution unsafe yaml.load() src/koru/context_sprint.py :23 return yaml.load(f, Loader=loader)
dynamic code execution unsafe yaml.load() src/koru/planfile_compat.py :86 document = yaml.load(stream, Loader=_yaml_loader())
dynamic code execution eval()/exec() test_openrouter_simple.py :56 exec(code, namespace)
dynamic code execution __import__ sink tests/test_ide_doctor_cli.py :375 assert "KORU_AUTOPILOT_INSTANCE" not in __import__("os").environ
dynamic code execution __import__ sink tests/test_loop.py :88 with patch("koru.loop.os.scandir", wraps=__import__("os").scandir) as scan:
dynamic code execution __import__ sink tests/test_vdisplay_env_session.py :20 assert "KORU_AUTONOMY_SESSION_DIR" not in __import__("os").environ

other grade factors

evidence elsewhere

recent drift +20 capability drift →

embed badge

readme-ready

[![MCP Observatory risk grade](https://mcpobservatory.com/servers/github:semcod/koru/badge.svg)](https://mcpobservatory.com/servers/github:semcod/koru/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 semcod.