Code analysis
static source read inferred
Static code-analysis findings — hidden prompt content in shipped skill files, committed secrets, dynamic-exec sinks, and suspicious call-home endpoints — across the analyzed catalogue. Heuristic, pure, no code executed; every row deep-links to its source. Click a kind to filter.
50100 analyzed
8073 re-analysis due
988 not analyzable
2 not yet analyzed
4804 source gone
not analyzable
789 too large 199 no source
Running analyzer v33. The scanner changelog explains what each version detects and when it changed.
- hidden prompt 338
- committed secret 6180
- dynamic exec 11839
- obfuscation 3598
- suspicious endpoint 12356
- credential in log 713
- over-broad oauth scope 2550
- suspicious skill script 185
- bundled IDE extension 49
- skill file 187702
- MEDIUM suspicious endpoint SK-DEV-AI/codesearch 1.1.1.1
for target, url in [("cloudflare", "https://1.1.1.1"), ("google", "https://www.google.com"), ("github", "https://api.github.com")]: - MEDIUM suspicious endpoint edward-arinin-web-dev/dexe-mcp t.me
["telegram", "https://t.me/fulldao"], - HIGH dynamic exec edward-arinin-web-dev/dexe-mcp new Function()
const makeValidate = new Function(`${names_1.default.self}`, `${names_1.default.scope}`, sourceCode); - HIGH dynamic exec uxarray-mcp __import__ sink
"slurm_job_id": __import__("os").environ.get("SLURM_JOB_ID"), - HIGH dynamic exec uxarray-mcp __import__()
module = __import__(name) - HIGH committed secret Luckycat133/skills-repo Anthropic key
sk-ant…(31 chars, redacted) - HIGH dynamic exec dannysheehan/ftmon __import__()
assert "response_time_s" not in __import__( - HIGH dynamic exec DHLbigmonster/CoEditor eval()
async eval(expression, { awaitPromise = true } = {}) { - HIGH committed secret sakebomb/mcp-recall Anthropic key
sk-ant…(46 chars, redacted) - HIGH committed secret sakebomb/mcp-recall OpenAI key
sk-svc…(61 chars, redacted) - HIGH dynamic exec sakebomb/mcp-recall new Function()
const makeValidate = new Function(`${names_1.default.self}`, `${names_1.default.scope}`, sourceCode); - MEDIUM over-broad oauth scope deepakkamboj/connected-workspace-mcp gmail.modify
'https://www.googleapis.com/auth/gmail.modify', - MEDIUM over-broad oauth scope deepakkamboj/connected-workspace-mcp gmail.modify
'https://www.googleapis.com/auth/gmail.modify', - MEDIUM suspicious endpoint Circe22/agents-guide-to-telegram api.telegram.org
TG_API = "https://api.telegram.org" - MEDIUM suspicious endpoint Circe22/agents-guide-to-telegram api.telegram.org
LEAK_URL = f"https://api.telegram.org/bot{FAKE_TOKEN}/sendMessage"