AI Agent Hub
Back to plugins
🤖

dsh-context-lens

Model Inference Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install 1014029855/dsh-context-lens

Paste the following prompt into your AI chat to install this plugin:

In DeepSeek Harness, you can install this plugin by running `dsh plugin install 1014029855/dsh-context-lens` in the command line. The source code is available at https://github.com/1014029855/dsh-context-lens.

About this plugin

When building AI agents powered by large language models, understanding the full context behind each model call is crucial. This goes beyond the messages and tool definitions sent to the model, encompassing how system prompts are assembled. However, this information is often buried in the underlying system, making it challenging for developers to grasp intuitively and leading to inefficient debugging and optimization.

The dsh-context-lens plugin addresses this need for DeepSeek Harness users by providing a transparent window. It adds a "Context" tab to sessions, visually displaying detailed information for each AgentLoop request. Developers can clearly see the model used, system prompts, message history, tool configurations, and the space occupied by each part, even comparing differences between requests.

The plugin's core capabilities lie in its meticulous observation and analysis. It accurately measures text metrics like UTF-8 bytes, UTF-16 code units, and provides estimated token counts, helping developers better understand context scale. Additionally, it supports raw and structured viewing, search functionality, and JSON export for deep diagnostics and issue sharing. All operations are read-only, ensuring no interference with the original request flow.

dsh-context-lens is particularly suited for developers and debuggers working with DeepSeek Harness. If you're building or optimizing AI agents and need to dive deep into model interaction details, this plugin will be an invaluable assistant. It enables quick identification of context issues, boosting development efficiency and ensuring the accuracy and reliability of agent behavior.

Use Cases

  • Debugging model request context for AI agents.
  • Analyzing the assembly process of system prompts and message history.
  • Evaluating context size and token overhead across different requests.

Best For

  • Engineers developing and debugging AI agents.
  • Researchers studying context usage in large language model applications.
  • Platform operators needing to monitor resource usage of agent systems.