aps-docs-portal
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches documentation structure (JSON) and content (HTML) from official Autodesk domains, specifically developer.doc.config.autodesk.com and developer.doc.autodesk.com. These are recognized as trusted vendor resources.
- [COMMAND_EXECUTION]: The skill utilizes curl, jq, and htmlq to retrieve and process documentation content. These tools are used for their intended purpose of data retrieval and structured parsing from official sources.
- [INDIRECT_PROMPT_INJECTION]: The skill ingest data from external HTML documentation pages to answer user queries. While the source is a trusted vendor, processing external text introduces a theoretical surface for indirect prompt injection.
- Ingestion points: Documentation URLs fetched in SKILL.md (Steps 3, 5, and 6).
- Boundary markers: None explicitly defined in the processing instructions.
- Capability inventory: Uses shell-based tools (curl, jq, htmlq) to fetch and parse data.
- Sanitization: Employs htmlq to strip script, style, and noscript tags from the HTML content before the agent processes the text, reducing the risk of processing hidden instructions.
Audit Metadata