skill-router
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a router that ingests untrusted user input during an interview process and interpolates that data into prompts intended for other skills.
- Ingestion points: User responses to 'Funnel Questions' (Step 2) in
SKILL.mdregarding their task, tech stack, and goals. - Boundary markers: Absent. The instructions do not specify using delimiters or 'ignore embedded instructions' warnings when generating the final prompt for the user.
- Capability inventory: None. This skill itself does not have file-write, network, or subprocess capabilities; it only outputs text recommendations.
- Sanitization: Absent. The skill does not explicitly sanitize or validate the user's input before incorporating it into the suggested prompts.
- [SAFE]: The skill is entirely instructional and does not contain any executable code, external dependencies, or network operations. It functions as a decision tree implemented through natural language prompts.
Audit Metadata