jinko-task-extract-data-table
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from potentially untrusted external sources (papers, web sources, images, PDFs) and possesses the capability to upload results to a server. This constitutes an indirect prompt injection surface where malicious instructions could be embedded in the source documents to influence the agent's behavior during extraction or upload.
- Ingestion points: Data is ingested from 'papers, figures, tables, supplements, images, or web sources' as defined in Step 1 of the Workflow in
SKILL.md. - Boundary markers: No explicit instructions are provided to the agent to ignore or delimit embedded instructions within the source documents.
- Capability inventory: The skill uses the
jinko-data-tablecommand to perform data uploads and interact with a server-side fitness function gate (Step 6 inSKILL.md). - Sanitization: There are no documented sanitization steps for the external content before it is processed or used to generate the CSV/Markdown output.
Audit Metadata