Skip to content
edisyledisyl

Search these docs from your AI tool

  • Antigravity
    {
  "mcpServers": {
    "edisyl-docs": {
      "serverUrl": "https://docs.edisyl.com/mcp"
    }
  }
}
  • Claude CodeCLI
    claude mcp add edisyl-docs https://docs.edisyl.com/mcp --transport http --scope user
  • CodexCLI
    codex mcp add --url https://docs.edisyl.com/mcp edisyl-docs
  • Cursor
    {
  "mcpServers": {
    "edisyl-docs": {
      "type": "http",
      "url": "https://docs.edisyl.com/mcp"
    }
  }
}
  • Gemini CLI
    gemini mcp add --transport http edisyl-docs https://docs.edisyl.com/mcp
  • Goose
    {
  "extensions": {
    "edisyl-docs": {
      "enabled": true,
      "name": "edisyl-docs",
      "type": "streamable_http",
      "uri": "https://docs.edisyl.com/mcp",
      "envs": {},
      "env_keys": [],
      "headers": {},
      "description": "",
      "timeout": 300,
      "bundled": null,
      "available_tools": []
    }
  }
}
  • JetBrains AI Assistant
    {
  "mcpServers": {
    "edisyl-docs": {
      "type": "http",
      "url": "https://docs.edisyl.com/mcp"
    }
  }
}

    Paste into Settings → Tools → AI Assistant → Model Context Protocol → Add → As JSON. Direct file writing isn’t supported — JetBrains stores this per-version as XML.

  • Junie (JetBrains)
    {
  "mcpServers": {
    "edisyl-docs": {
      "url": "https://docs.edisyl.com/mcp"
    }
  }
}
  • OpenCode
    {
  "mcp": {
    "edisyl-docs": {
      "type": "remote",
      "url": "https://docs.edisyl.com/mcp"
    }
  }
}
  • VS CodeCLI
    code --add-mcp '{"name":"edisyl-docs","type":"http","url":"https://docs.edisyl.com/mcp"}'
  • Windsurf
    {
  "mcpServers": {
    "edisyl-docs": {
      "serverUrl": "https://docs.edisyl.com/mcp"
    }
  }
}

    Supports both stdio and native HTTP connections.

Looking to ask a pack from your AI tool instead? That's a different server: see AI clients.

What is edisyl

AI rarely says “I don’t know.” It just answers (most confidently exactly when it’s wrong), and the cost shows up in the decision, not in the chat window.

edisyl develops Knowledge Packs: the context AI needs to make expert-grade decisions accurately and efficiently. A Knowledge Pack is a continuously current body of domain and proprietary knowledge, grounded in the way an organization actually works.

What goes into a Knowledge Pack

Section titled “What goes into a Knowledge Pack”

A pack combines two kinds of knowledge:

  • Domain knowledge: the leading edge. Constantly updating external knowledge (including licensed feeds), assembled by agents and refined by human experts. One body of domain knowledge can serve an entire vertical, or be refined further for a specific need.
  • Proprietary knowledge: an organization’s own documents, databases, and definitions, streamed in mostly through automated integrations. Stays private to that pack and makes the answers right for that specific context.

Domain knowledge is the foundation, built once and reused; proprietary knowledge layers on top of it, private to whoever owns the pack. A new question is usually a new pack built on that foundation, not a build from scratch.

Read the full breakdown in Knowledge Packs, including how the domain and proprietary halves combine.

The pack is the center of everything

Section titled “The pack is the center of everything”

A pack is the unit of:

  • durable understanding,
  • source attachment (knowledge, connections, data sources),
  • ownership and access,
  • provenance and freshness,
  • learning and curation,
  • work and activity history.

Every command in the CLI takes a pack as its target, and so does every tool on the MCP surface your AI client connects to. There’s no unscoped question, no unscoped fact, no unscoped work: everything is this pack’s knowledge, this pack’s task, this pack’s history.

Three actions, one mental model

Section titled “Three actions, one mental model”

People (and the CLI) interact with a pack through three actions:

  • Ask: answer a question from one or more packs, without changing them.
  • Teach (the CLI verb for what the product model calls “Learn”): add or correct durable understanding in exactly one pack.
  • Task: have one pack perform trackable work and produce a result.

These are contracts about user intent, not descriptions of runtime: an ask can take minutes, a task can finish in seconds. What matters is whether you expect an answer, a durable change, or delegated work.

Read the full breakdown in Ask, Teach, Task, or jump straight to Getting started.

Report incorrect code

Please provide a detailed description of the incorrect code.