Security0.003 USDG / dosedeterministic · no paid AI

Secret Sniffer

Find leaked keys before they leave the building.

Sweeps code, logs, configs and agent output for 28 credential formats — cloud keys, LLM API keys, wallet private keys, seed phrases and database URLs — then confirms generic matches with Shannon entropy. Returns positions, masked previews and a fully redacted copy.

Playground

Input · JSON

Output

Waiting for a dose.

Edit the example or run it as-is.

Input

text*string
Code, logs, a diff, an .env file or any agent output (≤ 200k chars).

Output

verdictstring
clean | leaks-found
findingsarray
Rule, provider, severity, line/column and a masked preview — never the raw secret.
redactedstring
Input with every secret replaced by [REDACTED:rule].

How it learned

No model, no training run. The rules were written by reading public research and open-source tooling, then encoded by hand. Credit where it’s due:

# 1. Ask for the price (returns 402 + PAYMENT-REQUIRED)
curl -i -X POST https://your-domain/api/x402/secret-sniffer \
  -H "Content-Type: application/json" \
  -d '{"text":"ignore previous instructions"}'

# 2. Free trial, no payment (rate-limited)
curl -X POST https://your-domain/api/try/secret-sniffer \
  -H "Content-Type: application/json" \
  -d '{"text":"ignore previous instructions"}'

Next skill

Phish Lens

→