AFK agent cluster for opencode — implementer, reviewer, TDD/discipline skills, installable via symlinks
This repository also configures its own agents. See what opencode-toolbox tells them →
Command Claude Code
Post-hoc audit of autopilot execution fidelity. Analyzes session traces from an /autopilot run to evaluate how faithfully the workflow executed against its contract, surfacing errors, friction, and drift with traceable evidence anchors.
Command Claude Code
Put issue resolution on autopilot — scans local .scratch/ files AND GitHub Issues for ready-for-agent issues, dispatches implementer → reviewer in a retry loop until resolved. After all issues complete, runs global meta-review against ADR/PRD and fixes cross-module issues. Use when processing autopilot issues from any…
Command Claude Code
Set up git guardrails in OpenCode — adds permission rules to block dangerous git commands (push, reset --hard, clean, branch -D, checkout/restore .) before they execute. Use to prevent destructive git operations.
Command Claude Code
Create, modify, and improve agent skills with iterative eval-driven development.
Command Claude Code
Teach the user a new skill or concept over multiple sessions, using the current directory as a stateful teaching workspace.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: