market-research
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted web data during its research phase, creating a potential surface for indirect prompt injection, which is typical for web-research tools.
- Ingestion points: External data from the internet enters the agent context through tools like WebSearch, WebFetch, and Firecrawl, as specified in the instructions for Layer 1 agents (e.g., agents/trends-agent.md and agents/competitor-agent.md).
- Boundary markers: The instructions for the research agents do not currently utilize explicit delimiters or instructions to ignore instructions that might be embedded in the external content.
- Capability inventory: The skill is configured to use the Bash tool and has permissions to write research reports to the local file system and update project-specific memory in the .agents/experience/ directory.
- Sanitization: No explicit sanitization or filtering of the content retrieved from external sources is implemented before it is processed by the analysis pipeline.
Audit Metadata