html-slide-prompt
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content from files like
DESIGN.mdandslide_plan.jsonto generate structured JSON output. This creates a surface for indirect prompt injection if those source files contain malicious instructions. - Ingestion points: Processes content from
DESIGN.md,slide_plan.json, and other source files as specified in theInputssection ofSKILL.md. - Boundary markers: The skill instructions do not define clear delimiters or use "ignore embedded instructions" warnings to separate user-provided data from the skill's operational logic.
- Capability inventory: The skill itself does not invoke dangerous tools, perform network operations, or execute shell commands; its primary function is data transformation and output generation to a JSON file.
- Sanitization: There is no evidence of content sanitization or validation of the ingested data to prevent it from influencing the agent's behavior.
Audit Metadata