qveris
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from a large variety of external providers (e.g., FMP, AlphaVantage, Finnhub) via the QVeris marketplace. This external content, which includes news, sentiment analysis, and financial filings, could be used for indirect prompt injection attacks where malicious instructions are embedded in the data stream to influence the agent's behavior.\n
- Ingestion points: Data returned by
qveris_executeand full payloads retrieved fromfull_content_file_url(described inSKILL.mdandreferences/rest-api.md).\n - Boundary markers: No specific delimiters or safety instructions are provided to help the agent distinguish between data and potential commands within the external payloads.\n
- Capability inventory: The skill utilizes the
qveris_executetool for marketplace operations and network requests to retrieve truncated content.\n - Sanitization: The documentation does not specify any sanitization, filtering, or validation processes for the data retrieved from external providers.\n- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch data from dynamic, signed URLs (
full_content_file_url) when the primary API response is truncated. Fetching content from dynamic external sources introduces a risk if the source is compromised or if the URL points to malicious content masquerading as data.
Audit Metadata