recon-laundromats

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses curl to perform network reconnaissance against target domains, checking for HTTP status codes and identifying CMS signatures.
  • [DATA_EXFILTRATION]: Instructions facilitate the discovery of sensitive information on target servers, such as customer PII in debug logs and exposed invoice files.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface when processing remote content:
  • Ingestion points: Fetches untrusted data from target websites using curl (SKILL.md).
  • Boundary markers: No delimiters are present to isolate remote data from the agent's instructions.
  • Capability inventory: Includes the ability to execute network requests, file system interactions (via curl output redirection), and data parsing with grep and jq.
  • Sanitization: The skill does not perform sanitization of the remote data before it is processed by shell utilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 05:19 PM
Security Audit — agent-trust-hub — recon-laundromats