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.
51193 analyzed
7507 re-analysis due
995 not analyzable
0 not yet analyzed
4848 source gone
not analyzable
795 too large 200 no source
Running analyzer v33. The scanner changelog explains what each version detects and when it changed.
- hidden prompt 369
- committed secret 6298
- dynamic exec 12007
- obfuscation 3647
- suspicious endpoint 12542
- credential in log 718
- over-broad oauth scope 2556
- suspicious skill script 197
- bundled IDE extension 50
- skill file 190152
- MEDIUM suspicious endpoint AlpayC/mcp-migration-check 172.15.0.1
["http://172.15.0.1/mcp", "just below 172.16.0.0/12"], - MEDIUM suspicious endpoint AlpayC/mcp-migration-check 93.184.216.34
["http://93.184.216.34/mcp", "public IPv4 literal"], - MEDIUM suspicious endpoint AlpayC/mcp-migration-check 169.254.169.254 (cloud metadata)
["http://169.254.169.254/latest/meta-data", "cloud metadata address"], - HIGH committed secret agent-airlock OpenAI key
sk-LIV…(39 chars, redacted) - MEDIUM suspicious endpoint shipmail-to/shipmail-mcp 8.8.8.8
expect(isPublicHttpsUrl("https://8.8.8.8/x")).toBe(true); - MEDIUM suspicious endpoint shipmail-to/shipmail-mcp 1.1.1.1
expect(isPublicHttpsUrl("https://1.1.1.1/x")).toBe(true); - MEDIUM suspicious endpoint shipmail-to/shipmail-mcp 169.254.169.254 (cloud metadata)
expect(isPublicHttpsUrl("https://169.254.169.254/latest/meta-data/")).toBe(false); - MEDIUM suspicious endpoint shipmail-to/shipmail-mcp 169.254.169.254 (cloud metadata)
url: "https://169.254.169.254/hook", - HIGH obfuscation peopleworks/SignsofAI dynamic require()/import()
const { narrate } = await import(`${KIT}/narrate.mjs`); - HIGH obfuscation peopleworks/SignsofAI dynamic require()/import()
const { narrate, getKey } = await import(`${KIT}/narrate.mjs`); - HIGH committed secret YoruichiYams/devguard-mcp OpenAI key
sk-pro…(40 chars, redacted) - HIGH credential in log howells/agentsurface credential in log
console.log("[client] Got token:", access_token); - MEDIUM suspicious endpoint akbarharyadi/claude-code-telegram api.telegram.org
url = f"https://api.telegram.org/bot{config.TELEGRAM_BOT_TOKEN}/{method}" - MEDIUM suspicious endpoint akbarharyadi/claude-code-telegram api.telegram.org
API_ROOT = "https://api.telegram.org" - MEDIUM suspicious endpoint akbarharyadi/claude-code-telegram api.telegram.org
url = f"https://api.telegram.org/bot{config.TELEGRAM_BOT_TOKEN}/getMe"