skills/bighardperson/computer-science-skills-collection/neodata-financial-search/Gen Agent Trust Hub
neodata-financial-search
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes external financial articles and market data from
copilot.tencent.com, creating an attack surface for instructions embedded in external content. - Ingestion points: Data enters the agent's context through the
apiRecall(structured data) anddocList(financial articles) fields returned by thescripts/query.pyandscripts/query.shscripts. - Boundary markers: The skill does not implement explicit delimiters or instructions for the agent to isolate and ignore potentially adversarial instructions within the retrieved financial data.
- Capability inventory: The skill is granted
BashandReadpermissions to execute its local querying scripts and read documentation. - Sanitization: The response data from the API is presented to the agent without text filtering or sanitization, relying on the agent's underlying safety guardrails.
Audit Metadata