research
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a CLI named 'story' to manage research notes and validate project state. It includes fallback instructions to execute local scripts using 'node' or 'bun' (e.g.,
node ../story-maintenance/scripts/story.js), which is standard for maintaining project continuity within this ecosystem. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize external information from web searches and user-provided documents.
- Ingestion points: Research findings obtained via web search, fetch tools, or user documents processed in
SKILL.md(Workflow section). - Boundary markers: None explicitly defined for research findings.
- Capability inventory: File system writes to
research/directory and shell command execution via thestoryCLI. - Sanitization: No specific sanitization or filtering logic is provided for external content before it is recorded in research notes.
Audit Metadata