quality-gate

Pass

Audited by Gen Agent Trust Hub on Nov 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from the repository and external sources to generate executable project infrastructure.
  • Ingestion points: Accesses project-local source code, configuration files, and 'current official documentation' from external sources during initialization and profile compilation (SKILL.md, Section 1 and 4).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the agent when processing external or repository-local data.
  • Capability inventory: The agent is authorized to modify or create critical project infrastructure, including GitHub Actions workflows ('.github/workflows/*'), test scripts, and static analysis configurations (SKILL.md, Section 5).
  • Sanitization: Lacks explicit sanitization or validation of content fetched from external documentation or project files before its inclusion in the prompt context or generated outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 20, 2026, 08:54 PM