aar
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional Markdown content and templates for debriefing events. It does not contain any executable scripts, network requests, or sensitive file path references.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided event details through its templates.\n
- Ingestion points: The skill ingests user data via the
$ARGUMENTSplaceholder inSKILL.md,commands/aar.md, andcommands/review.md.\n - Boundary markers: No specific delimiters or "ignore previous instructions" warnings are used to wrap the injected user data.\n
- Capability inventory: No tools (e.g., shell access, file system write, network) are enabled or utilized by the skill.\n
- Sanitization: No validation or sanitization of input data is performed.\n
- Note: While the skill provides an ingestion surface for untrusted data, the absence of exploitable capabilities or tool access renders the risk minimal.
Audit Metadata