Definitive cross-harness reference for configuring agents (rules/instructions, prompts, skills, agents/subagents, hooks, tools/MCP, plugins/extensions, secrets, sandbox, verification). Use to answer "how do I configure X?" with exact file paths, precedence, schemas, and verification steps.
EnglishExtract and organize Claude Code session history into project .chats directory. Use when users want to document their Claude sessions, export conversation inputs, or maintain a log of instructions given to Claude.
EnglishAnalyzes, generates, and enhances CLAUDE.md files for any project type using best practices, modular architecture support, and tech stack customization.
EnglishInspect, back up, and safely edit OpenCode agent/category model choices in…
EnglishUse when working with the standard GitHub CLI (gh) for repository collaboration workflows, especially opening, triaging, reviewing, merging, or closing GitHub pull requests. Apply when Codex needs an opinionated PR queue workflow with explicit review-comment retrieval, merge/close decisions, and supporting issue or PR creation commands.
EnglishUse for UI ideation, Google Stitch prompt packs, Stitch MCP design retrieval, DESIGN.md design-system creation, and implementing Stitch designs in a VS Code project. Use when asked to design screens, landing pages, dashboards, SaaS/mobile UI, convert Stitch output to frontend components, or keep UI consistent from Stitch artifacts.
EnglishOperates on the IcePanel REST API v1 for C4 landscapes — model import, positioned diagrams, flows, ADRs, drafts, landscape merge/copy, exports, share links, webhooks, and action-log forensics. Use when working with IcePanel / icepanel.io: importing a landscape, fixing a blank canvas, merging landscapes, or C4 modeling via the API or MCP.
EnglishReference guide for deploying and configuring MCP servers across all AI tools. Covers config file locations, format differences, secrets management, and deployment patterns for Claude Code, VS Code, Gemini CLI, OpenCode, Kilo Code, Antigravity, and other MCP-compatible clients.
EnglishExpert guide for the NotebookLM CLI (nlm) and MCP server - interfaces for Google NotebookLM.
EnglishWhen remediating phase N of a multi-phase plan, never assume phases 1..N-1 are satisfied — probe artifact existence on disk AND on the branch you claim to inherit from, AND confirm promotion (merge/PR) to the destination, before scoping phase N…
EnglishComplete reference for creating Claude Code plugins from scratch. Covers directory structure, plugin.json manifest, SKILL.md format, agents, hooks, MCP servers, commands, rules, marketplace publishing, and testing. Use when building a new plugin, adding skills to an existing plugin, or publishing a plugin marketplace.
EnglishFormats every agent chat reply with a restrained visual language — word-level color accents, status/severity/metric/entity palette, semantic emoji, safe Mermaid, progress bars, badges, and callouts. Apply by default to all responses; use richness dial L0 for one-line factual answers, L1+ when structure helps. General-purpose across domains (reviews, roadmaps, incidents, decisions, architecture). Skip only if the user asks for plain/minimal output.
EnglishInstall, configure, update, health-check, and switch model profiles for OpenCode and the oh-my-openagent (omo) plugin. Use when asked to install/upgrade/update OpenCode, check OpenCode/openagent versions, switch agents between OpenAI and cheap/non-OpenAI models, fix or edit oh-my-openagent.json, inspect which model each agent uses, or back up the OpenCode config.
EnglishAnalyze agentic coding sessions to discover friction patterns and generate skills, rules, and hooks. Use when running /insight across multiple sessions or optimizing AI assistant configuration. Do NOT use for single sessions or non-agentic tasks.
EnglishHelps set up universal rule templates in your project's .claude/rules/ directory. Explains each rule template and guides you through customizing them for your project.
EnglishAuto-selects and injects relevant skills from the 125+ ECC catalog into agent prompts based on task type, project stack, and agent role. Prevents skill overload by loading only what's needed.
EnglishInstall, consolidate, or repair Agent Skills (SKILL.md) across multiple AI coding tools on this machine using a single-source-of-truth layout. Use when asked to install a skill "in all coding tools", when a skill is duplicated across tool directories (~/.codex/skills, ~/.copilot/skills, etc.), when a skill shows up multiple times in a tool's picker, or when migrating skills to the .agents/skills open standard. Covers Claude Code, OpenCode (+oh-my-openagent), Codex, Gemini, Cursor, Copilot, and Kilo.
EnglishUse when converting messy branch or PR state into reviewable stacked pull requests with normal git and GitHub tooling. Applies when branches need semantic slicing, PR bases need retargeting, descendant branches need restacking, or checkpoint-heavy work must converge into a coherent bottom-up review story. Graphite gt can be used as an optional accelerator, but is not required.
EnglishIncludes Agent Config Megaref, Chat History, Claude Markdown Enhancer, and other workflows.
Supported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Agent Config Megaref
Definitive cross-harness reference for configuring agents (rules/instructions, prompts, skills, agents/subagents, hooks, tools/MCP, plugins/extensions, secrets, sandbox, verification). Use to answer "how do I configure X?" with exact file paths, precedence, schemas, and verification steps.
Chat History
Extract and organize Claude Code session history into project .chats directory. Use when users want to document their Claude sessions, export conversation inputs, or maintain a log of instructions given to Claude.
Claude Markdown Enhancer
Analyzes, generates, and enhances CLAUDE.md files for any project type using best practices, modular architecture support, and tech stack customization. Use when setting
Edit Opencode Config
Inspect, back up, and safely edit OpenCode agent/category model choices in oh-my-openagent.json
Gh Common Workflows
Use when working with the standard GitHub CLI (gh) for repository collaboration workflows, especially opening, triaging, reviewing, merging, or closing GitHub pull requests. Apply when Codex needs an opinionated PR queue workflow with explicit review-comment retrieval, merge/close decisions, and supporting issue or PR creation commands.
Google Stitch UI Designer
Use for UI ideation, Google Stitch prompt packs, Stitch MCP design retrieval, DESIGN.md design-system creation, and implementing Stitch designs in a VS Code project. Use when asked to design screens, landing pages, dashboards, SaaS/mobile UI, convert Stitch output to frontend components, or keep UI consistent from Stitch artifacts.
Icepanel API
Operates on the IcePanel REST API v1 for C4 landscapes — model import, positioned diagrams, flows, ADRs, drafts, landscape merge/copy, exports, share links, webhooks, and action-log forensics. Use when working with IcePanel / icepanel.io: importing a landscape, fixing a blank canvas, merging landscapes, or C4 modeling via the API or MCP.
MCP Deployment Guide
Reference guide for deploying and configuring MCP servers across all AI tools. Covers config file locations, format differences, secrets management, and deployment patterns for Claude Code, VS Code, Gemini CLI, OpenCode, Kilo Code, Antigravity, and other MCP-compatible clients.
Nlm Skill
Expert guide for the NotebookLM CLI (nlm) and MCP server - interfaces for Google NotebookLM. Use this skill when users want to interact with NotebookLM programmatically
Phantom Substrate Inheritance
When remediating phase N of a multi-phase plan, never assume phases 1..N-1 are satisfied — probe artifact existence on disk AND on the branch you claim to inherit from, AND confirm promotion (merge/PR) to the destination, before scoping phase N work
Plugin Authoring Guide
Complete reference for creating Claude Code plugins from scratch. Covers directory structure, plugin.json manifest, SKILL.md format, agents, hooks, MCP servers, commands, rules, marketplace publishing, and testing. Use when building a new plugin, adding skills to an existing plugin, or publishing a plugin marketplace.
Rich Visual Responses
Formats every agent chat reply with a restrained visual language — word-level color accents, status/severity/metric/entity palette, semantic emoji, safe Mermaid, progress bars, badges, and callouts. Apply by default to all responses; use richness dial L0 for one-line factual answers, L1+ when structure helps. General-purpose across domains (reviews, roadmaps, incidents, decisions, architecture). Skip only if the user asks for plain/minimal output.
Run Opencode
Install, configure, update, health-check, and switch model profiles for OpenCode and the oh-my-openagent (omo) plugin. Use when asked to install/upgrade/update OpenCode, check OpenCode/openagent versions, switch agents between OpenAI and cheap/non-OpenAI models, fix or edit oh-my-openagent.json, inspect which model each agent uses, or back up the OpenCode config.
Session Skill Inferencer
Analyze agentic coding sessions to discover friction patterns and generate skills, rules, and hooks. Use when running /insight across multiple sessions or optimizing AI assistant configuration. Do NOT use for single sessions or non-agentic tasks.
Setup Rules
Helps set up universal rule templates in your project's .claude/rules/ directory. Explains each rule template and guides you through customizing them for your project.
Skill Injector
Auto-selects and injects relevant skills from the 125+ ECC catalog into agent prompts based on task type, project stack, and agent role. Prevents skill overload by loading only what's needed.
Skill Install
Install, consolidate, or repair Agent Skills (SKILL.md) across multiple AI coding tools on this machine using a single-source-of-truth layout. Use when asked to install a skill "in all coding tools", when a skill is duplicated across tool directories (~/.codex/skills, ~/.copilot/skills, etc.), when a skill shows up multiple times in a tool's picker, or when migrating skills to the .agents/skills open standard. Covers Claude Code, OpenCode (+oh-my-openagent), Codex, Gemini, Cursor, Copilot, and Kilo.
Stacked PR Workflow
Use when converting messy branch or PR state into reviewable stacked pull requests with normal git and GitHub tooling. Applies when branches need semantic slicing, PR bases need retargeting, descendant branches need restacking, or checkpoint-heavy work must converge into a coherent bottom-up review story. Graphite gt can be used as an optional accelerator, but is not required.
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Agent Config Megaref skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- License details are not provided in these listings. Check the author’s terms before using the skills.