docs-draft-workflow-docs
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (
gh) to retrieve issue data. This usage is restricted to authorized vendor repositories (elastic/docs-contentandelastic/docs-content-internal) to facilitate the documentation intake process.\n- [EXTERNAL_DOWNLOADS]: The documentation instructions include the use ofnpx skillsto install companion tools from official vendor repositories. It also references an MCP server hosted on the vendor's official domain (elastic.co). These are recognized as legitimate vendor resources.\n- [DATA_EXFILTRATION]: The skill usesWebFetchand theghtool to access documentation and issue content. These network operations are strictly targeted at the vendor's infrastructure and are necessary for the skill's primary function.\n- [PROMPT_INJECTION]: The skill processes external data from GitHub issues which could contain untrusted instructions.\n - Ingestion points: Issue body, comments, and titles fetched via the GitHub CLI.\n
- Boundary markers: The skill employs a strict multi-step intake and classification workflow (Steps 0-6) which requires the agent to summarize and validate data before drafting.\n
- Capability inventory: The skill possesses file writing (
Write), command execution (Bash), and tool calling (CallMcpTool) capabilities.\n - Sanitization: The instructions provide specific YAML schemas and documentation templates that the agent must adhere to, reducing the risk of executing instructions embedded in the intake data.
Audit Metadata