research-ideation
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or suspicious instructions were found in the skill body or metadata.
- [DATA_EXFILTRATION]: The skill is designed to read from local directories ('master_supporting_docs/', '.claude/rules/') and write to 'quality_reports/'. These operations are limited to the agent's expected workspace and do not target sensitive system paths or include any network-based exfiltration mechanisms.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes user-provided topics and local documents. 1. Ingestion points: '$ARGUMENTS' variable and files in 'master_supporting_docs/'. 2. Boundary markers: Absent. 3. Capability inventory: 'Read', 'Grep', 'Glob', and 'Write' tools. 4. Sanitization: Filenames are sanitized using the '[sanitized_topic]' template. The risk is evaluated as safe because the skill lacks network access or shell execution capabilities required for data exfiltration or system compromise.
Audit Metadata