skills/dafang/codestable/cs-onboard/Gen Agent Trust Hub

cs-onboard

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several Python scripts that execute shell commands via subprocess.run. A notable example is tools/codestable-dod-runner.py, which executes commands specified in a local YAML checklist to verify project 'Definition of Done' criteria. This is a core feature of the workflow automation.
  • [EXTERNAL_DOWNLOADS]: The documentation suggests installing the @alibaba-group/open-code-review package from a well-known service provider. The skill correctly instructs the agent to seek user confirmation before performing this global installation.
  • [DATA_EXFILTRATION]: Tools within the skill, such as tools/build-review-packet.py and tools/codestable_common.py, are capable of reading sensitive project files (e.g., .env, credentials). These tools implement automated redaction to strip secrets and tokens before the content is shared or reviewed, adhering to security best practices for handling sensitive data within a repository.
  • [PERSISTENCE]: The skill provides scripts and configuration files to establish persistent hooks, including Git pre-commit hooks and agent-level PreToolUse hooks. These are intended to enforce branch protection policies (preventing direct edits on main) and are presented as optional components for the project owner to enable.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 12:38 PM
Security Audit — agent-trust-hub — cs-onboard