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.

analysis coverage 86% of 59,713 analyzable servers
51308 analyzed
7409 re-analysis due
996 not analyzable
0 not yet analyzed
4850 source gone
not analyzable
796 too large 200 no source

Running analyzer v33. The scanner changelog explains what each version detects and when it changed.

code findings 15 shown
  1. HIGH dynamic exec astrogilda/waitbus eval()/exec() exec(src, namespace)
  2. HIGH dynamic exec astrogilda/waitbus __import__ sink utime_ns, stime_ns = read_daemon_cpu_ns(pid=__import__("os").getpid())
  3. HIGH committed secret astrogilda/waitbus GitHub fine-grained PAT github…(93 chars, redacted)
  4. MEDIUM suspicious endpoint brainbytes-dev/leashd us.posthog.com (telemetry) host: process.env.NEXT_PUBLIC_POSTHOG_HOST || 'https://us.posthog.com',
  5. MEDIUM suspicious endpoint brainbytes-dev/leashd us.posthog.com (telemetry) api_host: process.env.NEXT_PUBLIC_POSTHOG_HOST || 'https://us.posthog.com',
  6. HIGH dynamic exec kakon77777-commits/eveaegis-mcp pickle.loads() restored = pickle.loads(pickle.dumps(make_grant()))
  7. MEDIUM over-broad oauth scope kakon77777-commits/eveaegis-mcp delete_repo if "delete_repo" not in scopes:
  8. HIGH dynamic exec mockworld-mcp eval()/exec() return bool(eval(expr, {"__builtins__": _SAFE_BUILTINS}, namespace)) # noqa: S307 - trusted local
    mockworld_mcp-0.3.2/src/mockworld/ faults.py :73
  9. HIGH committed secret JimmyWesley/MonkeyLLM Slack token xoxb-a…(24 chars, redacted)
  10. MEDIUM suspicious endpoint blinkboxhq/Blinkbox api.telegram.org const { data: fileInfo } = await axios.get(`https://api.telegram.org/bot${token}/getFile`, { params: { file_id: mediaFileId }, timeout: 10000 });
  11. MEDIUM suspicious endpoint blinkboxhq/Blinkbox api.telegram.org nt-mono text-[10px] text-zinc-500">POST https://api.telegram.org/bot&lt;TOKEN&gt;/setWebhook</span> with <span className="font-mono text-zinc-500">url</span> and optional <span className="font-mono te
  12. MEDIUM suspicious endpoint blinkboxhq/Blinkbox your-ngrok.ngrok.io placeholder="http://127.0.0.1:1234 or https://your-ngrok.ngrok.io"
  13. MEDIUM suspicious endpoint blinkboxhq/Blinkbox abc123.ngrok.io Then paste the ngrok URL (e.g. <span className="font-mono text-zinc-500">https://abc123.ngrok.io</span>) as Base URL below.{" "}
  14. MEDIUM suspicious endpoint blinkboxhq/Blinkbox api.telegram.org nt-mono text-[10px] text-zinc-500">POST https://api.telegram.org/bot&lt;TOKEN&gt;/setWebhook</span> with <span className="font-mono text-zinc-500">url</span> and optional <span className="font-mono te
  15. MEDIUM suspicious endpoint blinkboxhq/Blinkbox 1.1.1.1 await assert.doesNotReject(() => assertSafeUrlResolved("https://1.1.1.1/"));