qveris-official

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it ingests and processes tool metadata and results from the QVeris API tools.
  • Ingestion points: Untrusted data enters the agent's context from API responses fetched in scripts/qveris_client.mjs.
  • Boundary markers: The skill does not implement specific delimiters or instructions to the agent to ignore potentially malicious content within the tool results.
  • Capability inventory: The skill allows network requests to be made to the vendor's API.
  • Sanitization: Data is parsed as JSON, but the textual content is not sanitized for natural language instructions.
  • [SAFE]: The skill uses the QVERIS_API_KEY environment variable for authentication, which is a standard and secure practice for credential management.
  • [SAFE]: Outbound network traffic is limited to the official vendor domain (qveris.ai) via HTTPS.
  • [SAFE]: The implementation relies exclusively on built-in Node.js features (fetch, AbortController) and does not require external package installations or runtime script generation.
  • [SAFE]: Installation and persistence instructions are transparent and emphasize non-persistent session setup to avoid unauthorized system modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 09:05 AM