> For the complete documentation index, see [llms.txt](https://docs.cherryai.com.cn/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.cherryai.com.cn/docs/en-us/advanced-basic/developer-tools/trace.md).

# Call Chain and Developer Mode

The call chain is used to view the model requests, tool calls, and MCP interactions that occur within a conversation or Agent task. It is suitable for troubleshooting and performance analysis, and is not a feature that must be enabled for everyday chatting.

<figure><img src="/files/e1008326e4b5249f1b6cf4356bd2504258532d3c" alt="从运行环境、认证网络到 Agent 绑定和调用链的 MCP 排错流程图"><figcaption><p>When troubleshooting MCP calls, first confirm the server and authentication, then use the call chain to determine whether the request entered the Agent and tool execution stages.</p></figcaption></figure>

## How to enable

{% stepper %}
{% step %}

### 1. Open \[Settings] → \[System]

In the \[Developer Mode] section, enable \[Enable Developer Mode].
{% endstep %}

{% step %}

### 2. Restart Cherry Studio

Developer mode takes effect after restarting. Confirm the status of any Agent tasks currently running before restarting.
{% endstep %}

{% step %}

### 3. Reproduce the issue once

Resend the minimum test message, keeping only the necessary model and tools to reduce irrelevant calls.
{% endstep %}

{% step %}

### 4. Open the call chain

In the right-side panel of the conversation or Agent, enter \[Call Chain], select a node to view the input, output, duration, and status. Agent tasks can also show the Agent name, action, and number of tool calls.
{% endstep %}
{% endstepper %}

## How to read common nodes

| Node            | Key checks                                      | Common questions                                          |
| --------------- | ----------------------------------------------- | --------------------------------------------------------- |
| Model request   | Model, input, output, tokens, and status        | Provider errors, context too long, empty output           |
| Tool call       | Tool name, parameters, result                   | Parameter error, permission waiting, abnormal tool return |
| MCP call        | Service name, connection type, input and output | Server disconnect, authentication failure, remote error   |
| Agent execution | Agent, action, tool count, and status           | Subtask failed, process did not end                       |
| HTTP request    | Method, URL, status, and response               | Wrong address, network or authentication issue            |

{% hint style="danger" %}
Call chains may contain prompts, file contents, request headers, and tool parameters. Before sharing screenshots or exported information, you must remove API keys, Authorization, cookies, email addresses, local paths, and business data.
{% endhint %}

## User case: MCP tool returned an empty result

First confirm on the MCP settings page that the server is working normally, then have the Agent call only one tool. The call chain shows that the request reached the server but the output is empty, which means the problem is more likely in the server-side data or parameters, rather than the Agent not being bound; if there is no MCP node at all, go back to Agent \[MCP] to check the binding.

<details>

<summary>Why can't I see the call chain even after enabling it?</summary>

Confirm that you have restarted the app, and after restarting, initiate the request again. Historical messages generated before enabling will not automatically be backfilled with call data.

</details>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.cherryai.com.cn/docs/en-us/advanced-basic/developer-tools/trace.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.