compound-learnings

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources including git commit logs, pull request descriptions, and agent handoff files to identify patterns.
  • Ingestion points: The skill reads from git log, gh pr list, .agents/handoffs/*.md, and CLAUDE.md.
  • Capability inventory: The skill uses the Bash tool to analyze data and, upon approval, is designed to create or update project artifacts such as rules, skills, hooks, and agents.
  • Boundary markers: No specific delimiters or safety instructions are defined in SKILL.md to isolate the extracted pattern data from the agent's operational instructions.
  • Sanitization: The skill does not implement sanitization or filtering for the content extracted from commit messages or PR descriptions before recommending new artifacts based on those patterns.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute shell commands for repository analysis, specifically utilizing git log, grep, sort, and uniq to count and categorize recurring patterns in the codebase and history.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 12:28 AM