skill-authoring
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a local validation utility,
scripts/validate_skill.sh, which performs string manipulation and regex checks to ensure generated skills follow specific naming and metadata conventions. - [INDIRECT_PROMPT_INJECTION]: As a code generator, the skill ingests user input to produce new instruction files. It implements boundary checks via the validation script, which enforces a strict schema for the YAML frontmatter and limits the use of unsupported keys.
- [SAFE]: External links provided in the reference documentation point exclusively to official Claude Code and Anthropic documentation on well-known, trusted domains.
Audit Metadata