snapshot-writer
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external research data including paper abstracts, outlines, and goals to generate a summarized briefing. While this defines a potential surface for indirect prompt injection, the skill's capabilities are limited to local file operations within the workspace.
- Ingestion points: The script
scripts/run.pyreads data fromoutline/outline.yml,papers/core_set.csv,papers/papers_dedup.jsonl, andGOAL.md. - Boundary markers: No specific delimiters are implemented in the adapter script to isolate untrusted content.
- Capability inventory: The skill performs local file reading and writing within the provided workspace. No subprocess execution, system command calls, or network access patterns were identified.
- Sanitization: Input content is processed directly by the rendering tools without additional escaping or validation in the adapter layer.
Audit Metadata