skill-progressive-disclosure-design

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill metadata includes a project homepage URL (https://github.com/samber/cc-skills) that correctly maps to the author's GitHub profile and official repository.
  • [SAFE]: The content of SKILL.md consists exclusively of instructional prose and logic-based guidelines for context management and file splitting. It does not contain executable code, shell commands, or network-based exfiltration patterns.
  • [SAFE]: The evals/evals.json file contains 13 test cases designed to validate the agent's adherence to the skill's design principles. These evaluations do not contain malicious payloads or prompt injection attempts aimed at bypassing safety guardrails.
  • [SAFE]: The tool configuration in the frontmatter (allowed-tools) follows standard platform practices and does not grant excessive or unusual permissions beyond what is required for an assistant agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:18 AM
Security Audit — agent-trust-hub — skill-progressive-disclosure-design