writing-shape
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill behaves as intended for a document shaping tool. It performs read and write operations on local markdown files based on user direction. The instructions include safeguards like verifying the file content from disk before writing to prevent accidental data loss. No suspicious network activity, command execution, or credential exposure was found.
- [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for indirect prompt injection because it processes untrusted user-supplied markdown files. This is inherent to its primary purpose and carries no inherent risk beyond standard LLM input processing.
- Ingestion points: Reads a markdown file containing raw material provided by the user as described in
SKILL.md. - Boundary markers: None are present; the skill treats all material in the input file as content to be processed.
- Capability inventory: The skill includes file system read and append capabilities for content creation (
SKILL.md). - Sanitization: No explicit sanitization or input filtering is mentioned in the instructions.
Audit Metadata