Agent♥︎Age
Catalog

Contextful

Official

by Inferensys · TypeScript

Efficient context management: code search, evidence packs, and memory for coding agents.

Contextful MCP Server (io.github.Inferensys/contextful)

Contextful is an early-alpha runtime contextual layer and local search engine for agents. It provides a fast way to find, compress, cite, and remember project context, supporting code search, evidence packs, and memory for agentic AI. It is available as a CLI-first tool with an MCP runtime bridge and generated agent instructions.

🛠️ Key Features

  • Efficient context management
  • Code search
  • Evidence packs
  • Memory for agentic AI
  • MCP runtime bridge
  • Generated agent instructions

🚀 Use Cases

  • Supplying agents with project context
  • Helping agents locate relevant code and evidence
  • Supporting agent memory workflows
  • Improving token efficiency for coding agents

⚡ Developer Benefits

  • One fast way to find, compress, cite, and remember project context
  • Local context management + search
  • CLI-first workflow with MCP integration

⚠️ Limitations

  • Early alpha status
  • API, features, and integrations are not stable
  • Expect breaking changes

Topics

agent-memoryagentic-aiai-agentsclaude-codeclinecode-searchcodexcontext-managementcontinuecursorevidence-packsgithub-copilotmcpmodel-context-protocolroo-codetoken-efficiencyvscodewindsurfzed