latchshot-page-capture

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled Node.js script (scripts/latchshot.mjs) to perform web captures. The script is well-structured and uses built-in Node.js modules for file operations and networking.
  • [DATA_EXFILTRATION]: The skill intentionally sends public URLs to a specific hosted service (https://latchshot.fly.dev) for rendering. It includes multiple security checks to prevent sensitive information from being sent, such as:
    • Rejecting URLs with embedded credentials (username/password).
    • Restricting ports to standard web ports (80, 443).
    • Requiring explicit confirmation (--allow-query) before sending URLs with query strings to avoid accidental token exposure.
    • Restricting captures to public HTTP/HTTPS pages and explicitly rejecting local file access or internal network IPs.
  • [CREDENTIALS_UNSAFE]: The skill correctly handles the LATCHSHOT_API_KEY by reading it exclusively from the environment. It explicitly forbids passing the key as a command-line argument and includes redaction logic in its error handling to prevent the key from appearing in logs or console output.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 06:56 PM
Security Audit — agent-trust-hub — latchshot-page-capture