brand-context

Pass

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill creates a persistent data store (brand-context.md) by ingesting data from external URLs, which presents a surface for indirect prompt injection attacks.
  • Ingestion points: Content is ingested from user-provided URLs via the firecrawl_urls_scrape and web_scrape_page tools as described in SKILL.md.
  • Boundary markers: No delimiters or specific instructions are provided to the agent to treat the scraped content as untrusted or to ignore embedded instructions.
  • Capability inventory: The skill has file system write permissions via create_file and edit_file tools, as specified in the Tool surface section of SKILL.md.
  • Sanitization: The instructions do not define any validation or sanitization steps for the data retrieved from external websites before it is saved to the filesystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 8, 2026, 11:10 PM
Security Audit — agent-trust-hub — brand-context