irac-practice

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates entirely locally, managing student outlines and session history within a specific plugin configuration path (~/.claude/plugins/config/claude-for-legal/).
  • [PROMPT_INJECTION]: The skill ingests user essays, which are untrusted data. While this creates a surface for indirect prompt injection, the skill's capabilities are restricted to generating textual feedback and writing to a local session tracker, with no access to network operations or system commands to exploit.
  • [DATA_EXFILTRATION]: The skill reads from and appends to local files within its own configuration directory. There are no network operations or attempts to transmit data to external domains.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 03:58 PM
Security Audit — agent-trust-hub — irac-practice