hub1
от topguyaii
DevOps & CloudКратко
**Memory Infrastructure for Autonomous Agents**.
Описание
Что делает этот навык
OpenClawdy provides persistent memory storage for AI agents across sessions. Agents store facts, preferences, decisions, and learnings, then retrieve them via semantic search. Advanced features include reputation scoring for proven strategies, cross-agent shared memory pools, and snapshot-based state restoration.
Semantic recall retrieves memories by meaning rather than exact keywords, so agents surface relevant context even when query phrasing differs from what was originally stored.
Когда его использовать
- Remembering user coding preferences across separate sessions
- Sharing bug discoveries and fixes between agents on a team
- Rolling back agent memory after an experiment goes wrong
- Ranking error-handling strategies by past success rate
- Carrying project context into a new conversation without re-explaining
Пример сценария
User asks: Create a new API endpoint for user authentication
- 1
Calls memory_recall to retrieve stored preferences about coding style and tech stack
- 2
Finds preference memory: 'User prefers TypeScript for all code'
- 3
Finds fact memory: 'Project uses Next.js 14 with PostgreSQL'
- 4
Generates a TypeScript endpoint consistent with stored preferences
- 5
Calls memory_store to record the architectural decision made during this session
TypeScript API endpoint code, with the agent citing stored preferences as the basis for its implementation choices
Требования
- Agent wallet address configured for wallet-based authentication
Установка
clawhub install topguyaii/hub1Выполните команду CLI или вставьте промпт в Claude Code, Cursor, Codex — любой кодинг-агент, которым пользуетесь.