project-continue
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests data from external project files to determine its next development steps, creating a surface for indirect prompt injection.
- Ingestion points: Reads PRDs from
.claude/blueprints/prds/, feature tracker fromdocs/blueprint/feature-tracker.json, and work orders from.claude/blueprints/work-orders/. - Boundary markers: Absent. The skill does not use XML tags or specific delimiters to separate documentation content from instructions.
- Capability inventory: Has access to
Bash,Write,Edit, andReadtools, enabling it to execute commands and modify files based on ingested content. - Sanitization: Absent. No validation or filtering of content from the project files is performed before processing.
Audit Metadata