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.
50066 analyzed
8097 re-analysis due
988 not analyzable
0 not yet analyzed
4803 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 6178
- dynamic exec 11834
- obfuscation 3579
- suspicious endpoint 12346
- credential in log 713
- over-broad oauth scope 2550
- suspicious skill script 185
- bundled IDE extension 49
- skill file 187628
- HIGH dynamic exec serhii-nochevnyi/shipyard new Function()
const wf = new Function('agent', 'parallel', 'phase', 'log', 'args', - HIGH obfuscation serhii-nochevnyi/shipyard dynamic require()/import()
const { acquire, sleepSync } = require(${JSON.stringify(LOCK)}); - MEDIUM suspicious endpoint aidvgg/n8n-mcp api.telegram.org
url: "=https://api.telegram.org/file/bot{{ $credentials.telegramApi.accessToken }}/{{ $json.result.file_path }}", - HIGH committed secret EastStarAI/sanad-agent OpenAI key
sk-g6c…(29 chars, redacted) - HIGH obfuscation aeonfun/aeon dynamic require()/import()
await import(new URL(`../fleet-scorecard.mjs?test=${Date.now()}`, import.meta.url)) - HIGH committed secret aeonfun/aeon OpenAI key
sk-DRY…(39 chars, redacted) - MEDIUM suspicious endpoint aeonfun/aeon api.telegram.org
await fetch(`https://api.telegram.org/bot${env.TELEGRAM_BOT_TOKEN}/answerCallbackQuery`, { - MEDIUM suspicious endpoint aeonfun/aeon mcp.posthog.com (telemetry)
url: 'https://mcp.posthog.com/mcp', - HIGH committed secret aeonfun/aeon GitHub fine-grained PAT
github…(42 chars, redacted) - MEDIUM suspicious endpoint aeonfun/aeon api.telegram.org
const getUpdatesUrl = trimmed ? `https://api.telegram.org/bot${trimmed}/getUpdates` : null - MEDIUM suspicious endpoint aeonfun/aeon t.me
href="https://t.me/BotFather" - MEDIUM suspicious endpoint aeonfun/aeon t.me
<a href="https://t.me/BotFather" target="_blank" rel="noopener noreferrer" - MEDIUM suspicious endpoint aeonfun/aeon api.telegram.org
`curl "https://api.telegram.org/bot${botToken.trim() || '<YOUR_BOT_TOKEN>'}/setWebhook?url=${fullWorkerUrl || 'https://<your-worker>.workers.dev'}"` - HIGH hidden prompt aeonfun/aeon skill: skill-exfil
secret→sink: 9. **Build payload + send.** Build the JSON with `python3` reading `RESEND_FROM`/`RESEND_REPLY_TO` from `os.environ` — so no secret-named var ever lands on a command line (a `--arg from " - HIGH hidden prompt aeonfun/aeon skill: skill-exfil
secret→sink: - POST to `https://api.resend.com/emails`: