architecture-compass

Pass

Audited by Gen Agent Trust Hub on May 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill analyzes untrusted repository files, creating a surface for indirect prompt injection where malicious code comments could influence the agent's behavior.
  • Ingestion points: Processes up to 25 files from the repository, including dependency manifests (e.g., package.json, pom.xml) and source code entry points.
  • Boundary markers: No specific delimiters or 'ignore embedded instructions' warnings are used when the agent reads external code files.
  • Capability inventory: The skill is capable of reading repository files and writing a summary insights document to .lattice/insights/architecture.md.
  • Sanitization: Extracted code signals and file contents are not sanitized before being summarized into the final report.
Audit Metadata
Risk Level
SAFE
Analyzed
May 26, 2026, 11:04 PM
Security Audit — agent-trust-hub — architecture-compass