orbitant-learning-insights

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing unstructured text from previous interactions to build learner context and update insights.
  • Ingestion points: Historical lesson data in ~/.claude/learning/lessons/archive/ and session logs in ~/.claude/learning/logs/ (SKILL.md).
  • Boundary markers: Absent. The instructions do not define explicit delimiters or instructions to ignore embedded commands within the processed log files.
  • Capability inventory: The skill performs file-read operations on session history and file-write operations to the persistent insights.md file.
  • Sanitization: No evidence of sanitization, filtering, or validation of the ingested historical content is mentioned to prevent instruction leakage from past logs into the current session context.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 11:50 PM
Security Audit — agent-trust-hub — orbitant-learning-insights