knowledge-locator
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the retrieval of information from external sources such as memory palaces and a 'review chamber' containing past decisions and patterns, which represents an indirect prompt injection surface.
- Ingestion points: Untrusted data enters the agent context via search results from
scripts/memory_palace_cli.pytargeting memory palaces and project review repositories (documented inSKILL.md). - Boundary markers: There are no explicit instructions, delimiters, or 'ignore' directives provided to the agent to isolate potentially malicious instructions within the retrieved content.
- Capability inventory: The skill includes search and discovery capabilities through a command-line interface (
scripts/memory_palace_cli.py) and references external skill dependencies likereview-chamber. - Sanitization: The instructions do not define any sanitization, escaping, or validation steps for the retrieved palace data before it is processed by the agent.
Audit Metadata