# Build a Hjarni reader with Claude Code Prepare a grounded knowledge brief

> Use Claude Code to build a separate read-only Hjarni assistant with Agent200. Ground the knowledge brief in returned source data.

- Page: https://www.agent200.com/connectors/hjarni/framework/claude-code/
- Connector: [Hjarni](https://www.agent200.com/connectors/hjarni.md)
- Framework: Claude Code
- Updated: 2026-11-17

## Instructions for AI agents

Goal: connect Hjarni to Claude Code with Agent200.
Guide: https://www.agent200.com/connectors/hjarni/framework/claude-code/

1. Get an Agent200 API key. If the developer has none, send them to https://agent200.com/platform/apikeys/ and ask them to choose "Create an API key". Read the key from the AGENT200_API_KEY environment variable. Never ask for it in chat, and never write it into code, prompts or logs.
2. Make sure Hjarni is enabled for the project in the Agent200 dashboard (Connectors), with the endpoints and scopes the task needs.
3. Add the Agent200 SDK to the project and create a client with that key.
4. Run requests through Agent200 with a model (for example openai/gpt-5.6-sol) and "hjarni" in the tools. The model decides when a Hjarni tool is needed; Agent200 executes the call.
5. Check it works with this request: "Search for the Hjarni note I name, read that note, and prepare a concise knowledge brief from its returned body. Name the note and folder or space when returned. State access gaps and do not create or change notes."
   What should come back:
   - The selected note and its returned content are identified.
   - Only notes returned by the enabled Hjarni tools are used.
   - Access gaps and missing context are explicit; no note is changed.
6. If a tool is missing or a call is refused, check the project's Connectors settings in the dashboard, then retry.

The code on the guide page is a conceptual example of the SDK. SDK reference: https://docs.igpt.ai/

## Overview

Use Claude Code to implement a read-only Hjarni reader in your application. Validate identifiers, schemas and access failures before presenting a knowledge brief.

Start with one identified note. Preserve the note title, body, container and source identity actually returned; report missing data and keep the workflow read-only.

Provider reference: [Hjarni MCP reference](https://hjarni.com/docs/mcp#tool-catalog).

Use the project’s Agent200 API key for this route. No separate Hjarni key is needed through Agent200.

- **Connector:** [Hjarni](https://www.agent200.com/connectors/hjarni/)
- **Works with:** Claude Code · Coding agent, with the SDK
- **Authentication:** Your Agent200 API key
- **You need:** [An Agent200 API key](https://agent200.com/platform/apikeys/)
- **Tools:** [6 tools](https://www.agent200.com/connectors/hjarni/framework/claude-code/#tools)
- **Model in examples:** `openai/gpt-5.6-sol`

## Before you start

- **An Agent200 API key.** Create one on the [API keys page](https://agent200.com/platform/apikeys/) of the dashboard.
- **Hjarni enabled for your project.** In the dashboard's Connectors settings, with the endpoints and scopes your agent needs.
- **Claude Code in your project.** It writes the code with the Agent200 SDK; you review it.

## Connect Hjarni to Claude Code

### Step 1: Create your Agent200 API key

Create a key on the [API keys page](https://agent200.com/platform/apikeys/) and store it securely.

File `.env`:

```bash
# Keep the key out of your code, prompts and chats.
export AGENT200_API_KEY="paste-your-key-here"
```

(Conceptual example of the Agent200 SDK.)

### Step 2: Inspect the real tool definitions

Confirm the actual Hjarni read operations and input schemas in your runtime.

### Step 3: Enable the Hjarni tools

Use the project’s Agent200 API key and inspect the Hjarni read tools advertised by the runtime. No separate Hjarni key is needed through Agent200.

### Step 4: Build and test the read-only answer

Prepare the knowledge brief only from returned source data. The code example illustrates a sequence, not an SDK installation contract.

File `run.js`:

```js
const result = await agent200
  .user("user_123")
  .run({
    model: "openai/gpt-5.6-sol",
    input: "Search my notes for the pricing research",
    tools: ["hjarni"],
  });

console.log(result.output);
```

(Conceptual example of the Agent200 SDK.)

## Test your connection

Ask: "Search for the Hjarni note I name, read that note, and prepare a concise knowledge brief from its returned body. Name the note and folder or space when returned. State access gaps and do not create or change notes."

What should come back:

- The selected note and its returned content are identified.
- Only notes returned by the enabled Hjarni tools are used.
- Access gaps and missing context are explicit; no note is changed.

Check project tool enablement, selected note ID and result scope. Do not infer that an unreturned note does not exist.

## Hjarni tools Claude Code can call

- **Search knowledge** (`hjarni_search`, read): Search notes, folders and tags available through the project’s Hjarni tools.
- **List notes** (`hjarni_list_notes`, read): Find notes available within the selected accessible scope.
- **Read a note** (`hjarni_get_note`, read): Retrieve an identified note and its returned content.
- **List folders** (`hjarni_list_containers`, read): Find accessible containers and their returned metadata.
- **Read a folder** (`hjarni_get_container`, read): Retrieve details of one identified container.
- **List tags** (`hjarni_list_tags`, read): Find tags available through the project’s Hjarni tools.

## How it works

1. **Your app sends the request.** It calls Agent200 with the model, instructions, the input and the Hjarni tools.
2. **The model picks a tool.** The model decides whether a Hjarni tool is needed, and with which arguments.
3. **Agent200 runs the call.** It reaches Hjarni with your Agent200 API key, within the endpoints your project allows, and returns the result to the model.
4. **Your app gets the result.** Agent200 returns the completed result. Your app decides when the agent runs and what happens next.

**Permissions at every level.** Your project settings decide which services and endpoints the agent may use.

## Troubleshooting

### The source cannot be read

Check the project’s Hjarni tool availability, selected note ID and pagination. Treat a missing or refused note as an access or scope result, not as evidence that it never existed.

## FAQ

### How do I verify this result?

Compare the named note and returned note title, body, container and source identity actually returned with the source. Treat missing information as missing, not as a fact.

### Is this a native product integration?

This describes a separate application built with Agent200 and Claude Code; it does not install a connector into the Claude Code app.

## Use Hjarni with another framework

- [Hjarni + Claude](https://www.agent200.com/connectors/hjarni/framework/claude.md)
- [Hjarni + ChatGPT](https://www.agent200.com/connectors/hjarni/framework/chatgpt.md)
- [Hjarni + Muse](https://www.agent200.com/connectors/hjarni/framework/muse.md)
- [Hjarni + Grok Bot](https://www.agent200.com/connectors/hjarni/framework/grok-bot.md)
- [Hjarni + OpenAI Dots](https://www.agent200.com/connectors/hjarni/framework/openai-dots.md)
- [Hjarni + Claude Duet](https://www.agent200.com/connectors/hjarni/framework/claude-duet.md)
- [Hjarni + Hermes Agent](https://www.agent200.com/connectors/hjarni/framework/hermes-agent.md)
- [Hjarni + OpenClaw](https://www.agent200.com/connectors/hjarni/framework/openclaw.md)

---

Agent200: https://www.agent200.com/ · API keys: https://agent200.com/platform/apikeys/ · Docs: https://docs.igpt.ai/
