cerebrun
par niyoseris
Coding Agents & IDEsRésumé
MCP client for Cerebrun - comprehensive personal context and memory management system.
Description
Ce que fait ce skill
Cerebrun is a personal context and memory management system accessed via MCP. It stores user preferences, projects, goals, and knowledge across sessions in layered tiers, including an encrypted vault. An LLM Gateway lets agents route chat requests through configured providers.
Unlike session-only memory, Cerebrun persists context across different agents and tools without requiring manual re-injection each time.
Quand l'utiliser
- Retrieving your language and timezone preferences before starting a task
- Searching past notes for a specific project or decision
- Storing a new insight or idea to your personal knowledge base
- Looking up API keys or identity info stored in your context layers
- Routing a chat request through your preferred LLM provider via the gateway
Exemple de workflow
User asks: 'Find my notes on Rust authentication and save a new insight about JWT expiry handling'
- 1
Call search_context with query 'Rust authentication' to retrieve relevant past entries
- 2
Review returned results to identify existing notes
- 3
Call push_knowledge with content about JWT expiry handling, category 'learning', tags ['rust','auth']
- 4
Confirm the new entry was stored successfully
Returns matching Rust authentication notes and confirms the new JWT insight is saved to the knowledge base
Prérequis
- Cerebrun API key from cereb.run
- CEREBRUN_API_KEY environment variable
Installation
clawhub install niyoseris/cerebrunExécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.