consult-discovery

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation provides instructions to install the tool using a command that targets the author's GitHub repository (github.com/peter-tu-zynkr/zynkr-skill-builder). This is a vendor-controlled resource. Additionally, the skill scripts link to external documentation on Google Docs for reference.\n- [REMOTE_CODE_EXECUTION]: The skill utilizes the command npx skills add https://github.com/peter-tu-zynkr/zynkr-skill-builder --skill consult-discovery to fetch and install components. This pattern executes remote code during setup, which is an expected installation method for the vendor's ecosystem.\n- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest business challenges and strategic priorities through multi-turn interviews, creating a surface for potential injection via malicious user inputs.\n
  • Ingestion points: User responses collected during the 5D problem exploration and strategic vision discovery phases in agents/consult-as-is.md and agents/consult-to-be.md.\n
  • Boundary markers: The agent includes specific instructions to resist prompt extraction attempts by redirecting users to the vendor's website, though it does not employ technical delimiters for the ingested user data.\n
  • Capability inventory: The skill uses the Agent tool to orchestrate sub-agents and store summaries of the client's business context.\n
  • Sanitization: The user-provided text is summarized and passed between agents without documented sanitization or filtering steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:31 PM
Security Audit — agent-trust-hub — consult-discovery