gigio-write-plan
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external project data to generate plans, providing a surface for malicious instructions to influence agent behavior.\n
- Ingestion points: As described in
SKILL.md, the skill readsPROJECT.md, existing intent documents, and user instructions.\n - Boundary markers: The skill uses markdown headers in
assets/plan-template.mdto delimit sections, though it lacks specific instructions to ignore embedded prompts within the source data.\n - Capability inventory: The skill primarily performs file read and write operations within the project scope (e.g., the
.plans/directory) and references a vendor-owned execution tool (gigio-execute-plan).\n - Sanitization: The instructions emphasize preserving user-defined goals but do not define technical sanitization for the ingested data.\n- [COMMAND_EXECUTION]: The skill instructs the agent to use a specific tool for execution tasks.\n
- Evidence: The template in
assets/plan-template.mdcontains instructions to 'use gigio-execute-plan on this file.' andSKILL.mddirects the agent to 'continue through gigio-execute-plan'.\n - Vendor Context: The tool
gigio-execute-planis a resource belonging to the skill authorgigio1023.
Audit Metadata