mdapi-conversion

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits user-provided content such as URLs, files, and raw text to the vendor's API at mdapi.io for processing. This is the core functionality of the skill; however, users should ensure they do not unintentionally process sensitive local files like environment configurations or private keys.
  • [PROMPT_INJECTION]: The skill provides an interface to process untrusted data from external URLs and files using a user-supplied prompt, which constitutes a surface for indirect prompt injection. \n
  • Ingestion points: The url, file, and text parameters defined in SKILL.md. \n
  • Boundary markers: The instructions do not specify any delimiters or safety markers to help the agent distinguish between instructions and ingested data. \n
  • Capability inventory: The agent utilizes network requests to mdapi.io to transmit and receive converted data. \n
  • Sanitization: No explicit sanitization or filtering steps for external content are detailed in the skill instructions.
  • [CREDENTIALS_UNSAFE]: The skill instructions describe the use of authentication tokens and payment memos transmitted via HTTP headers or query parameters. No hardcoded credentials are included in the skill; it uses standard placeholders for user configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 10:40 PM
Security Audit — agent-trust-hub — mdapi-conversion