decision-council

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it automatically scans and ingests content from workspace files to provide context for its sub-agents.
  • Ingestion points: The agent is instructed to use Glob and Read tools to find and ingest content from CLAUDE.md, memory/ folders, and files containing pricing or revenue data.
  • Boundary markers: The instructions lack specific delimiters or instructions for sub-agents to ignore or isolate commands that might be embedded within the project files.
  • Capability inventory: The skill uses the Agent tool to spawn multiple sub-agents and has access to file system tools (Read, Glob, Write).
  • Sanitization: There is no defined process for sanitizing or validating the contents of the read files before they are interpolated into the sub-agent prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 06:23 PM
Security Audit — agent-trust-hub — decision-council