writing-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and review content from the
ideas/,drafts/, andpublished/directories, which provides a surface for indirect prompt injection via user-controlled files. - Ingestion points: Markdown files located in the project's drafting folders as described in
SKILL.md. - Boundary markers: The instructions do not define any delimiters or system-level warnings to distinguish user content from agent instructions during the review process.
- Capability inventory: The skill utilizes the
writetool to create and modify documents, and triggers metadata updates based on document content. - Sanitization: No explicit sanitization, escaping, or instruction-filtering is specified for the data processed from these files.
- [NO_CODE]: The skill consists entirely of markdown instructions and a JSON configuration file. No scripts, binaries, or automated installation procedures are included.
Audit Metadata