brownfield-integration

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to ingest and trust content from existing files on disk, creating a surface for indirect prompt injection where malicious instructions embedded in those files could influence agent behavior.
  • Ingestion points: The 'Read-Before-Write Checklist' in SKILL.md requires reading the full content of existing code files.
  • Boundary markers: The instructions lack directives for using delimiters or 'ignore embedded instructions' warnings when processing external file content.
  • Capability inventory: The agent is authorized to modify and write files to the local filesystem based on the ingested content.
  • Sanitization: There are no requirements for sanitizing or validating the patterns or comments read from existing files before they are adopted into the agent's plan or output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 05:49 PM
Security Audit — agent-trust-hub — brownfield-integration