00-andruia-consultant

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted local data by scanning workspace files (e.g., package.json, source code) during its technical diagnostic phase. This creates a surface for indirect prompt injection, where malicious instructions hidden in a project's files could influence the agent's behavior.
  • Ingestion points: Technical scanning of the workspace (src, package.json, etc.) in the 'Escenario B: Proyecto Existente' workflow.
  • Boundary markers: The instructions lack specific delimiters or guidelines for the agent to ignore instructions found within analyzed data.
  • Capability inventory: The skill is configured to use developer tools (e.g., claude-code, codex-cli) and performs file-writing operations to generate technical artifacts like tareas.md.
  • Sanitization: No explicit sanitization or content validation steps are defined for the ingested files before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 09:22 PM
Security Audit — agent-trust-hub — 00-andruia-consultant