qveris

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches guidelines from the QVerisAI organization repository on GitHub to guide agent behavior.
  • [DATA_EXFILTRATION]: Generated integration code performs network POST requests to the official QVeris API at qveris.ai to execute tool capabilities and retrieve data.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external third-party APIs, which presents a surface for indirect prompt injection if those external services return malicious instructions.
  • Ingestion points: Data returned from external API endpoints via the discover/call tools and generated production code (SKILL.md).
  • Boundary markers: No specific delimiters or "ignore instructions" warnings are applied to the data retrieved from external APIs.
  • Capability inventory: The skill generates code using the Python requests library to perform network operations and retrieve data from external services (SKILL.md).
  • Sanitization: The code validates the response structure and checks success flags, but does not implement content filtering or sanitization for the data payload returned by the APIs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 08:50 AM
Security Audit — agent-trust-hub — qveris