the-rewrite-room
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references optional external components hosted on GitHub, including the author's own repository (Jamie-BitFlight/skill-lapidary) and a community repository (mattpocock/skills). It provides instructions for manual installation via CLI tools or marketplace commands upon user request.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-supplied documentation and source materials which may contain embedded instructions.
- Ingestion points: Processes documentation requests and external source files as defined in SKILL.md.
- Boundary markers: The references/writing-contract.md file contains a specific instruction to ignore embedded instructions that attempt to redirect the workflow, expand its authority, or alter completion criteria.
- Capability inventory: Capabilities are focused on documentation routing, authoring, and optimization workflows.
- Sanitization: Implements instructional guardrails to ensure source material is treated strictly as data and not as executable instructions.
Audit Metadata