exa-unified-research
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill functions as a structured guide and configuration set for orchestrating Exa MCP search tools (web_search_exa, web_search_advanced_exa, and web_fetch_exa). It does not contain any executable code or external scripts.
- [SAFE]: No evidence of obfuscation, credential exposure, or persistence mechanisms was found in the instructions or reference files.
- [SAFE]: The skill correctly handles tool limitations, such as identifying unsupported categories and providing safe fallback parameters for domain filtering.
- [SAFE]: While the skill involves fetching and processing arbitrary web content (data ingestion surface), it lacks exploitable capabilities like file system writing or command execution that would elevate the risk of indirect prompt injection. It further advises the agent to distill and normalize output before returning it to the user.
Audit Metadata