build-guide

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to read from and write to the local file system. Specifically, it reads user role data from ~/.claude/plugins/config/claude-for-legal/legal-clinic/CLAUDE.md and writes new practice-area configuration files to the ~/.claude/plugins/config/claude-for-legal/legal-clinic/guides/ directory.- [PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by persisting user-provided instructions into configuration files that are subsequently ingested by other skills to control their pedagogy and review logic.
  • Ingestion points: User input captured during the 'Build Guide' interview process (Steps 2-7 in SKILL.md).
  • Boundary markers: The skill does not explicitly instruct the agent to use boundary markers or 'ignore' delimiters when writing the user-provided content into the markdown guide files.
  • Capability inventory: The skill's primary capabilities involve file system read and write operations within the plugin's configuration path.
  • Sanitization: No validation or sanitization of the user-provided text is performed before it is written to the guide files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 03:58 PM
Security Audit — agent-trust-hub — build-guide