skf-rename-skill

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill performs legitimate local filesystem management tasks. No evidence of data exfiltration, obfuscation, or persistence mechanisms was detected.\n- [COMMAND_EXECUTION]: The skill utilizes shell commands including cp -a, mv, and rm -rf to manage directories. These operations are restricted to the skill's defined output folders and are protected by explicit user confirmation and validation of the new name against a strict regex (^[a-z][a-z0-9-]*[a-z0-9]$).\n- [PROMPT_INJECTION]: The skill processes untrusted content from existing skill files, creating an indirect prompt injection surface (Category 8).\n
  • Ingestion points: Reads SKILL.md (frontmatter and body), metadata.json, and .export-manifest.json from the local filesystem.\n
  • Boundary markers: Recognizes YAML frontmatter delimiters (---) and managed section markers (<!-- SKF:BEGIN -->) to distinguish between structured data and free-text body.\n
  • Capability inventory: Includes filesystem manipulation (copying, renaming, deleting) and file write operations.\n
  • Sanitization: Validates the user-provided new name against format and length constraints before using it in filesystem paths or file contents.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 05:05 PM
Security Audit — agent-trust-hub — skf-rename-skill