api-doc-integration-risk-review

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits user-provided materials (API documentation text, uploaded file data, and external URLs) to the vendor's API at https://ai-skills.ai. This is the intended functional behavior of the skill for performing risk assessments.
  • [EXTERNAL_DOWNLOADS]: The Python runner script (scripts/run.py) establishes outbound HTTPS connections to https://ai-skills.ai to facilitate skill execution and retrieve analysis results.
  • [PROMPT_INJECTION]: The skill processes untrusted external input, creating a surface for indirect prompt injection attacks:
  • Ingestion points: Untrusted data enters the agent context through the apiGuideText, apiGuideFile, and apiGuideLink parameters defined in SKILL.md and references/skill.json.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided configuration.
  • Capability inventory: The skill executes network operations (HTTPS POST) to transmit data to the backend API via scripts/run.py.
  • Sanitization: There is no evidence of input validation or content filtering to mitigate malicious instructions hidden within processed documentation.
  • [COMMAND_EXECUTION]: The skill provides a Python script (scripts/run.py) intended for CLI execution to interact with the remote service API.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:31 AM
Security Audit — agent-trust-hub — api-doc-integration-risk-review