create-implementation-plan
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input via the
${input:PlanPurpose}variable to generate implementation plans. These plans are explicitly designed for autonomous execution by other AI systems, which creates a surface for indirect prompt injection if the generated content is interpreted as instructions by downstream agents. - Ingestion points: The user input variable
${input:PlanPurpose}is used as the basis for the generated plan in SKILL.md. - Boundary markers: The skill lacks explicit delimiters or warnings within the generated template to prevent downstream agents from following malicious instructions embedded in the user input.
- Capability inventory: The skill allows the agent to write generated Markdown files to the
/plan/directory. - Sanitization: There is no evidence of sanitization, escaping, or validation of the input content before it is interpolated into the template.
Audit Metadata