fittings-doc-html-to-markdown

Warn

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The documentation instructs users to download a remote JavaScript file from https://fittings.sh/skill/call.mjs and execute it using node, which constitutes unverified remote code execution.
  • [CREDENTIALS_UNSAFE]: The skill requires a sensitive wallet private key (FITTINGS_PRIVATE_KEY) to be stored in plaintext environment variables for payment settlement, posing a risk of credential theft.
  • [EXTERNAL_DOWNLOADS]: The skill fetches a script from the vendor domain and references third-party packages @x402/core and @x402/evm.
  • [COMMAND_EXECUTION]: Instructions involve running shell commands (curl, node) to interact with the conversion service.
  • [PROMPT_INJECTION]: The skill processes untrusted HTML, creating an indirect prompt injection surface. 1. Ingestion points: html parameter in POST request body. 2. Boundary markers: None specified. 3. Capability inventory: Network access to fittings.sh. 4. Sanitization: Claims to strip scripts and styles but lacks specific prompt injection protections.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 17, 2026, 10:31 PM