recap
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads and summarizes data from
.maestro/sessions/and.maestro/decisions.jsonl. This creates a surface for indirect prompt injection where instructions embedded in previous session logs or decisions could influence the agent's behavior during the recap process. - Ingestion points:
.maestro/sessions/and.maestro/decisions.jsonl(referenced in SKILL.md). - Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided for the ingested data.
- Capability inventory: The skill is primarily designed for data summarization; no subprocess calls, file-write operations, or network operations are present in this skill file.
- Sanitization: No explicit sanitization, validation, or filtering of the external log content is described.
Audit Metadata