review-skill-improver

Pass

Audited by Gen Agent Trust Hub on May 26, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill is focused on data analysis and generating recommendations to improve other skills. It follows a clear, instructional logic for aggregating feedback, identifying trends, and proposing modifications.
  • [COMMAND_EXECUTION]: The skill documents a usage pattern using a CLI-style invocation (/review-skill-improver) to output its analysis. This is a standard method for triggering specific agent workflows.
  • [INDIRECT_PROMPT_INJECTION]: While the skill is designed to process external feedback logs (an untrusted data surface), its capabilities are limited to generating structured markdown reports. It lacks the network or file-writing permissions necessary to turn a data injection into a significant security threat.
  • [METADATA_CONSISTENCY]: The skill's metadata (name, description, and user-invocable status) accurately reflects the content and purpose described in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 26, 2026, 05:53 AM
Security Audit — agent-trust-hub — review-skill-improver