brand-dna-builder
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill scrapes and processes external website content, which could be used to inject malicious instructions into the agent's context.
- Ingestion points: The skill fetches the homepage and subpages (such as /about or /products) using WebFetch or the Firecrawl tool (SKILL.md, Steps 2 and 3).
- Boundary markers: No specific delimiters are used to wrap the scraped content, although the skill follows a specific extraction checklist (references/extraction-checklist.md).
- Capability inventory: The skill utilizes the firecrawl_scrape tool and has the ability to write files to the local directory at
./brand/brand-dna.md(SKILL.md, Step 6). - Sanitization: The skill implements a robust mitigation by presenting all extracted data to the user and requiring explicit confirmation before any data is saved to the filesystem (SKILL.md, Step 4).
- [EXTERNAL_DOWNLOADS]: The skill initiates network requests to external domains provided by the user.
- It utilizes Firecrawl, a well-known service for web scraping, as a fallback when standard fetching methods are blocked (SKILL.md, Step 2).
- The skill provides clear instructions for the user to securely manage their own API credentials via platform settings.
Audit Metadata