YouTube MCP for Claude & ChatGPT
TubeLab은 Claude와 ChatGPT용 YouTube MCP 서버로, 실제 채널·바이럴 영상·전사 등을 AI 채팅에서 제공합니다
저장된 프로젝트 규칙·결정을 코딩 에이전트에 적용하고, 컨텍스트·런타임·재사용 스킬을 moldea와 맞춤
Use saved project rules and decisions in coding-agent planning and development. Keep context, runtime agents, and reusable skills aligned with moldea.
스킬TypeScript · JavaScriptGitHub @moldea-ai가동 상태 재확인 필요운영자 미확인
써 보셨어요? 한 번만 눌러 주세요. 로그인은 필요 없어요.
Official Agent Skill for building and continuously maintaining Git-native moldea project context and AI agent behavior.
README 전체를 불러오는 중입니다.
README 전문 보기 ↗Get moldea on skills.sh or read the complete documentation at skill.moldea.ai.
This checkout contains moldea 10.1.0. Install the latest published version from main inside each repository that will use it:
npx skills add moldea-ai/skill
For a reproducible installation, pin the immutable release after publication:
npx skills add "moldea-ai/skill#v10.1.0"
Both commands install the portable skill named moldea. They do not install the CLI globally or require a hosted account. See the Setup reference for initialization, updates, and removal.
Updating the skill does not update a project's CLI. Request explicit moldea repair or update the project's CLI development dependency to 13.0.4 through its package manager to receive YAML diagnostic locations.
moldea ismoldea helps developers keep durable project knowledge in Git and bring relevant context into coding-agent planning and development. Start with project purpose, rules, and decisions, even when your application has no runtime agents. The same foundation supports creating and maintaining runtime agents, reusable Agent Skills, and their instructions.
Adopted projects own a canonical /moldea/** tree. Focused context, implementation relationships, runtime guidance, mirrors, and unresolved requirements are added only when current project evidence calls for them.
The skill helps a coding agent:
After initialization, ask naturally to plan, build, review, or maintain AI agents, including model instructions and responsibilities in one-shot calls and workflows. Established project facts, approved policies, clear corrections, and project-context questions also receive bounded assessment without naming moldea or a file. Ordinary implementation requirements still use declared relationships, even when their effects will persist. When ordinary work expands to newly discovered paths, the coding agent checks those paths once at an existing scope checkpoint and preserves earlier matched owners. Only explicit initialization creates a new setup.
| Outcome | Example request |
|---|---|
| Initialize | Initialize moldea |
| Use context | Use the saved API rules to plan the next change. |
| Plan | Plan the agents for this project's support-triage workflow. |
| Create an agent | Create a support agent grounded in the current project policy. |
| Maintain context | Update moldea context for the approved refund policy. |
| Evaluate | Evaluate the current moldea project. |
| Reconcile | Reconcile the billing agent with its declared implementation. |
| Repair | Fix moldea. |
| Validate | Validate moldea. |
Evaluation is read-only. Repair checks every owner, its material implementation relationships, and each agent's instruction connection. It corrects established errors and updates local CLI tooling that is missing, broken, older than the exact target recorded by the installed skill, or without a supported root declaration. A large repair runs in rounds within the same conversation. Each round shows its progress, the files it changed and what remains, then offers numbered next steps; reply to continue. An incomplete pass is never called healthy. Repair preserves host restrictions, healthy newer compatible tooling, and unresolved policy. Validation checks declared files and explicit TypeScript or JavaScript ESM exports. Coverage names what was checked and does not prove runtime wiring. Warnings do not invalidate a project: ask to fix them or explicitly accept a supported warning with your own reason. The coding agent never chooses acceptance on your behalf.
여기까지가 README 앞부분입니다. 나머지는 원문에서 읽을 수 있습니다. README 전문 보기 ↗