recall
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a high-level orchestration instruction for rebuilding working context. It utilizes standard version control tools (git/gh) in a read-only capacity to verify state.
- [SAFE]: The instructions include explicit safety boundaries, such as 'Lock the scope' and 'Sanitize private context before any public output,' which reduce the risk of data leakage.
- [SAFE]: The skill references standard development tools and trusted upstream sources (GitHub) without initiating dangerous commands or remote code execution.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from chat histories and the 'shared record.' While this constitutes an attack surface, the instructions include boundary markers (locking scope) and sanitization steps. Given these mitigations and the primary purpose of context retrieval, the risk is minimal.
Audit Metadata