gaps
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill processes data from a local YAML file (
gap-tracker.yaml). This creates an attack surface where instructions embedded within the gap data could potentially influence agent behavior if the source file is modified by untrusted processes. - Ingestion points:
~/.claude/plugins/config/claude-for-legal/regulatory-legal/gap-tracker.yaml(SKILL.md) - Boundary markers: None specified in the instructions for isolating data from commands.
- Capability inventory: File read and write operations on the specified tracker file.
- Sanitization: No explicit sanitization of YAML content is mentioned.
- Local File Interaction: The skill is configured to read and modify a specific file path within the user's home directory. This behavior is transparently documented and aligns with the tool's primary purpose of gap tracking and remediation management.
Audit Metadata