cold-call-prep

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • Configuration Access: The skill retrieves personalized data, such as class lists and learning styles, from a specific local path (~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md). This allows the agent to tailor the Socratic drilling to the user's specific academic context.
  • Indirect Prompt Injection Surface: The skill processes untrusted input in the form of user-provided case text and reading assignments. (1) Ingestion points: Untrusted data enters via the user's input of case text, citations, or reading list paths specified in the skill's arguments. (2) Boundary markers: The skill does not employ specific delimiters or instructions to ignore embedded commands within the processed case text. (3) Capability inventory: The instructions in SKILL.md do not require or provide network access, file-writing capabilities, or shell command execution tools. (4) Sanitization: No explicit sanitization or filtering of input text is performed. The lack of sensitive capabilities mitigates the risks typically associated with this ingestion surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 07:01 PM
Security Audit — agent-trust-hub — cold-call-prep