nws-point-metadata

Warn

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: MEDIUMCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill instructions require the user to set the FITTINGS_PRIVATE_KEY environment variable with a raw Ethereum private key to authorize API payments. The documentation explicitly warns that this key remains in plaintext within the agent's execution environment, posing a risk of exposure if the host is compromised.
  • [EXTERNAL_DOWNLOADS]: The skill requires downloading a core execution script, call.mjs, from the vendor's domain (https://fittings.sh/skill/call.mjs). Executing remotely sourced scripts is a common attack vector, even when the source is the stated vendor.
  • [COMMAND_EXECUTION]: Operation of the skill involves running a Node.js script that performs dynamic imports and manages network communications. The script includes an automated integrity check that compares a local SHA-256 hash against a remote hash file (https://fittings.sh/skill/call.mjs.sha256), effectively enforcing remote updates.
  • [DATA_EXFILTRATION]: The execution script generates and transmits cryptographic signatures based on the user's private key to the fittings.sh gateway to process payment settlements for the service.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 18, 2026, 05:17 AM