Skills
skills/alirezarezvani/claude-skills/self-improving-agent/Gen Agent Trust Hub

self-improving-agent

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of processing untrusted data.
  • Ingestion points: The skill reads MEMORY.md and associated topic files in ~/.claude/projects/, which contain records of arbitrary command outputs and conversation history that may be influenced by external project files or data.
  • Boundary markers: Instructions in agents/memory-analyst.md and skills/review/SKILL.md do not implement explicit delimiters or escaping for the content being analyzed from the memory files.
  • Capability inventory: The skill includes commands to write to sensitive project configuration files (CLAUDE.md and .claude/rules/) via skills/promote/SKILL.md and generates new skill files via skills/extract/SKILL.md.
  • Sanitization: There is no evidence of sanitization or validation of the content extracted from memory before it is proposed as a permanent project rule.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 09:03 AM