remind
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves arbitrary content from a local database that may contain untrusted data from source files or user notes.
- Ingestion points: Content enters the agent's context through tools like MemoryFetch, MemoryFetchBatch, MemorySearch, and MemoryDelta (SKILL.md, references/snapshots-diffs.md).
- Boundary markers: The skill lacks instructions or delimiters to isolate retrieved database content from the agent's core instructions (SKILL.md).
- Capability inventory: The agent environment typically includes file-system access and shell execution capabilities, making it a target for instructions embedded in recalled memory (SKILL.md).
- Sanitization: Content is returned as raw text or structured data without sanitization against malicious prompts (SKILL.md, references/resources.md).
- [SAFE]: The skill references documentation on the author's repository for configuration details (references/resources.md).
Audit Metadata