code-search
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEDATA_EXFILTRATIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill provides search patterns for sensitive file paths such as
.envfiles to find credentials like passwords and keys. - [PRIVILEGE_ESCALATION]: Documentation includes examples of using
sudofor installing theripgreputility via system package managers likeaptandpacman. - [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to ingest untrusted data from local files which may contain embedded instructions.
- Ingestion points: File content read by the
ripgreptool as described inSKILL.md. - Boundary markers: No explicit markers or delimiters are used in the provided search command examples to distinguish data from instructions.
- Capability inventory: The skill utilizes
BashandTasktools which have the ability to execute shell commands. - Sanitization: No evidence of data sanitization, escaping, or instruction filtering is present in the search workflows.
Audit Metadata