ask-for-tables-over-paragraphs
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown documentation and instructional prompt templates. No scripts, binaries, or executable logic are included in the repository.
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates designed to process user-provided data (e.g., product names, metrics, categories).
- Ingestion points: Placeholders like
[ITEM A],[ITEM B], and[CATEGORY / TOOL]inSKILL.mddefine where external data is interpolated into the prompt. - Boundary markers: The instructions use structural constraints (e.g., "Format your response EXCLUSIVELY as a single Markdown table") to guide the AI, though these are for formatting rather than security isolation.
- Capability inventory: The skill has no active capabilities (no file system access, network access, or command execution).
- Sanitization: Not applicable; the skill is a collection of static text templates.
Audit Metadata