write-spec
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of implementation plans based on user-provided feature descriptions, which introduces a surface for indirect instructions. This risk is addressed through iterative clarification and review.\n
- Ingestion points: Feature descriptions provided by the user during the
Interviewphase (Workflow Step 1) and information gathered duringResearch(Workflow Step 2).\n - Boundary markers: The skill does not define specific syntax delimiters for user input but relies on structured interviewing to isolate requirements and define non-goals.\n
- Capability inventory: The skill writes to the
specs/directory, spawns subagents for parallel drafting, and performs web-based research.\n - Sanitization: The
Grill before planningprinciple requires the agent to validate outcomes with the user, and theRecursive fog auditdecomposes complex tasks to identify and isolate hidden variables.\n- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to perform 'targeted online research' into official documentation and reference implementations. These activities are limited to gathering information and do not involve the automated execution of remote scripts or unverified binary downloads.
Audit Metadata