replay-learnings
Pass
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No malicious patterns, suspicious commands, or safety guideline violations were detected in the skill instructions.
- [NO_CODE]: The skill consists entirely of Markdown-based instructions and metadata. It does not contain any executable scripts, shell commands, or binary files.
- [PROMPT_INJECTION]: The skill processes past interactions and user tasks, creating a surface for indirect prompt injection. Ingestion points: task keywords and session history in SKILL.md. Boundary markers: Absent. Capability inventory: The skill has no access to subprocesses, network requests, or file-writing tools. Sanitization: Absent. Due to the lack of executable capabilities, the risk associated with this ingestion surface is negligible.
Audit Metadata