soia-dev-coding-protocol

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation includes a setup command npx skills add soia-team/soia-open-dev-coding-skills which installs additional capabilities from the vendor's repository.
  • [COMMAND_EXECUTION]: The protocol instructs the agent to execute project-specific verification commands, such as linters, type checks, and test suites, to provide evidence of correctness before declaring a task complete.
  • [SAFE]: The skill contains explicit security constraints for the agent, such as:
  • Warning against concatenating untrusted input into shell commands.
  • Requiring explicit user authorization for remote operations like pushing code, publishing, or deleting resources.
  • Instructions to minimize code changes to prevent the introduction of unnecessary complexity or side effects.
  • [SAFE]: The execution contract emphasizes 'verifying results, not intent,' requiring actual command output and logs rather than subjective claims of success.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 01:39 PM
Security Audit — agent-trust-hub — soia-dev-coding-protocol