knowledge-accumulator
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of Markdown templates and methodology guides. No executable scripts, binary files, or installation manifests are included.
- [SAFE]: The skill defines a clear, manual folder structure and workflow for accumulating research data into specific files. All instructions align with legitimate knowledge management practices.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and store data from external sources such as web documentation, manuals, and codebase excerpts. 1. Ingestion points: corpus.md and source_map.md. 2. Boundary markers: Absent for content within the corpus. 3. Capability inventory: File write/read access and tool-based research. 4. Sanitization: Absent. The risk is mitigated by the skill's focus on source mapping and human-readable organization.
Audit Metadata