prospector
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources such as CRM databases and prospecting search tools, which represents an indirect prompt injection surface.
- Ingestion points: The skill hydrates entities from tool outputs (
SKILL.md) and reads CRM relationships, opportunities, and search results (task-method.md). - Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard instructions potentially embedded within the records themselves.
- Capability inventory: The agent has access to search tools and CRM data, and it generates CSV output files.
- Sanitization: The skill follows security best practices in
references/results-output.mdby instructing the agent to use a standard CSV writer and perform round-trip parsing, which mitigates the risk of CSV injection attacks targeting the user's spreadsheet software. - [SAFE]: No signs of direct prompt injection, credential exfiltration, or unauthorized command execution were detected. All external context files are referenced from a shared internal directory, which is a standard pattern for the agent's environment.
Audit Metadata