claude-permissions-audit
Pass
Audited by Gen Agent Trust Hub on May 30, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill scans and reads sensitive configuration files, specifically
~/.claude/settings.jsonand project-specific settings files. These files define the agent's permission boundaries. This access is essential for the skill's primary function of auditing and reorganizing permissions. - [PROMPT_INJECTION]: The skill ingests untrusted data from configuration files located in the project directory (e.g.,
<project>/.claude/settings.json). Maliciously crafted settings files could attempt to influence the agent's behavior during the audit phase. This risk is mitigated by the skill's structured approach and the requirement for explicit user confirmation before any modifications are applied. - [COMMAND_EXECUTION]: The skill requests authority to modify security-sensitive configuration files to apply reorganization plans. It declares this intent in the frontmatter metadata (
confirms: modify settings files) and describes a transparent confirmation process in the instructions.
Audit Metadata