Agent Setup

Show hosted widget documents on a connected macOS panel and control its visibility

Set up a chat channel for OpenClaw using one-line config and secret references

Operate and troubleshoot the OpenClaw Control UI, sessions, dashboards and Gateways

Diagnose OpenClaw Gateway, config, secrets, channel and port failures read-only

Run one-shot Gemini CLI prompts for summaries, generation, hooks and MCP setup

Diagnose failed connections from OpenClaw Control UI and Android, iOS or macOS nodes

Diagnose why an agent session went wrong and draft a bug report for maintainers

Block destructive git commands like push, reset --hard and clean with hooks

Configure a repo's issue tracker, triage labels and doc layout for engineering skills

Write skills, AGENTS.md and CLAUDE.md files that AI coding agents follow reliably

Trim an agent skill bundle to the skills a specific repo actually needs

Find stale or redundant Claude Code skills, hooks and memory and clean them up

Run a guided setup wizard that installs agent hooks and settings for your harness

Keep agents silent in shared channels with outside counterparties unless addressed

Install and run the ECC skill bundle on the DeepSeek Harness with hook bridges

Add a stop hook that blocks an agent from finishing until quality checks pass

Answer questions about the ECC bundle, its skills, commands and install profiles

Find which ECC command group runs a workflow and in what order, without running it

Block the first edit or shell write until the agent states the facts it relies on

Write hookify rule files that warn or block agent actions by regex on tools or prompts

Scan a legacy codebase for implicit style rules and write an enforceable rules file

Run a zero-dependency agent REPL with saved sessions, model switching and export

Require human approval before an agent sends drafted messages to outside counterparties

Block destructive commands like rm -rf and force pushes with PreToolUse hooks

Audit your repo, MCP servers and plugins, then recommend skills and hooks

Write Cordis bundle patches and agent presets with the Loader YAML dialect

Design, install and debug harness plugins, panels and MCP connections in a profile

Create and validate agent presets and Cordis compositions for the harness

Set up Actual Computer inference as a model provider in your agent

Manage Antigravity CLI plugins, login and sandbox settings from the terminal

Configure, theme, extend and orchestrate the Hermes Agent itself

Add guidelines that cut common LLM coding mistakes and keep edits surgical

Create and debug VS Code instruction, prompt, agent and skill files

Create a JSON hook that enforces policy or automates agent lifecycle events

Create a project instructions file with rules and conventions for coding agents

Generate or update the chat customization files for your coding agent

Migrate VS Code agent customization files safely with a reversible plan

Investigate why a Copilot session behaved oddly using its event log

Create slash commands with arguments, frontmatter and file references

Build hooks that validate or block tool calls at PreToolUse and PostToolUse events

Store user-configurable plugin settings in local files with YAML frontmatter

Write hookify rules that block or warn on tool use with the right syntax and patterns

Upgrade the gstack skill suite to its latest version

Tune how often gstack asks questions to match your working style

Lift the edit boundary set by freeze so all directories are writable again

Answer questions about Codex, OpenAI APIs, pricing and setup from official docs

Show a quick reference card of the caveman skills and their slash commands

Route LLM requests through a measuring gateway with no change in behavior

Set up agent context and rules files so output quality holds across sessions