konnect-platform-router

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected.
  • [PROMPT_INJECTION]: The skill contains standard instructional logic for task classification. There are no attempts to override safety filters, extract system prompts, or bypass agent constraints.
  • [DATA_EXFILTRATION]: The skill does not access sensitive local files (e.g., credentials, SSH keys) or perform network operations. It mentions the kong-konnect MCP server for live state, which is a platform-integrated tool for the intended product.
  • [REMOTE_CODE_EXECUTION]: There are no remote script downloads, piped execution patterns (e.g., curl|bash), or dynamic code evaluation.
  • [COMMAND_EXECUTION]: While the skill references CLI tools like kongctl, decK, and Terraform, these are listed as categories for routing purposes and are not invoked directly within the instructions.
  • [OBFUSCATION]: The content is clear and uses standard markdown and YAML. No encoded strings, zero-width characters, or hidden payloads were found.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 09:44 PM
Security Audit — agent-trust-hub — konnect-platform-router