apollo-company-research
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and reference files do not contain any malicious patterns, obfuscation, or unauthorized command execution sequences.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data from the Apollo API, including news and job postings. This creates a surface for indirect prompt injection where instructions could be embedded in the retrieved content. However, the skill provides clear guidance on synthesizing findings with evidence and maintaining attribution, which helps the agent distinguish between data and instructions.
- [COMMAND_EXECUTION]: The skill uses a defined interface (
sandbase_call_tool) to interact with specific Apollo research capabilities. This is a restricted execution model that limits the agent to the provided research tools.
Audit Metadata