commit-push
241Command Claude Code
Validate changes, commit with 'progress', and push to remote.
Command Claude Code
Validate changes, commit with 'progress', and push to remote.
Command Claude Code
Install live-bench — prerequisites, dependencies, models, and environment for the dashboard server and benchmark client machines.
Command Claude Code
Creates a concise engineering implementation plan based on user requirements and saves it to specs directory.
Command Claude Code
Load foundational context about the live-bench codebase.
Skill Claude CodeCodex
Terminal automation CLI for AI agents via tmux. Use drive to create tmux sessions, execute commands, send keystrokes, read output, poll for patterns, run commands in parallel across sessions, manage processes, and take screenshots of terminal windows. Always use --json for structured output. Triggers on terminal…
Instructions file
Claude Code instructions for disler/live-bench, covering claude.md, rules, benchmarks must never auto-reset and stack reminders.
Agent Claude Code
Documentation scraping specialist. Use proactively to fetch and save documentation from URLs as properly formatted markdown files.
Command Claude Code
Build the codebase based on the plan.
Command Claude Code
Load documentation from their respective websites into local markdown files our agents can use as context.
Command Claude Code
Creates a concise engineering implementation plan based on user requirements and saves it to specs directory.
Command Claude Code
Understand the codebase and the project.
Command Claude Code
Understand the codebase, project and operate on the sandboxcli.
Command Claude Code
Quickly prime yourself on the obox (Out Box Agent) command-line tool and sandbox fork workflow system. Learn how to run parallel agent forks in isolated E2B sandboxes for experimentation and testing.
Settings file Claude Code
Agent settings configuring version, settings.
Instructions file
Claude Code instructions for disler/agent-sandboxes.
Command Claude Code
Complete plan and build workflow - creates plan then implements it.
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: