crusty-old-engineer
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- Data Ingestion and Potential for Indirect Prompt Injection: The skill is designed to process untrusted data from multiple sources, including local project files via directory exploration tools (Read, Grep, Glob) and external web content via retrieval tools (WebFetch).
- Ingestion Points: Data enters the agent's context through user-provided code, architectural proposals, and fetched web pages.
- Boundary Markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded within the analyzed content.
- Capability Inventory: The skill utilizes information retrieval capabilities to examine codebases and search the web.
- Sanitization: There is no explicit requirement for the validation or filtering of content retrieved from external files or websites.
- External Resource Fetching: As part of its core behavior to provide evidence-linked judgment, the skill retrieves content from external technical documentation. It specifically targets established engineering resources such as the Google SRE Book and the AWS Builders' Library to anchor its recommendations in industry best practices and postmortem analyses.
Audit Metadata