spaced-repetition
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill architecture presents a potential attack surface for indirect prompt injection by design.\n
- Ingestion points: The system is built to process and display translation pairs and phrase data, which may come from untrusted external sources (specified in the Data Model section of SKILL.md).\n
- Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore' markers to separate untrusted phrase content from the agent's instructions.\n
- Capability inventory: The agent is authorized to use 'Read', 'Edit', 'Write', 'Grep', and 'Glob' tools (defined in the SKILL.md frontmatter), which provide a significant capability set for interaction with the local environment based on processed data.\n
- Sanitization: Absent. The documentation provides no guidance on sanitizing or validating ingested phrase text to prevent execution of embedded instructions.
Audit Metadata