dsh-yogacara
Run the following command in DeepSeek Harness:
dsh plugin install tancheng33/dsh-yogacara
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, use the command dsh plugin install tancheng33/dsh-yogacara, with the source code available at https://github.com/tancheng33/dsh-yogacara.
About this plugin
Intelligent agents often lack self-awareness during task execution, leading to repetitive errors, stubbornness, or unconscious loops, while simple mood labels like 'frustrated' offer no actionable regulation. The dsh-yogacara plugin builds a self-model based on Yogācāra philosophy, mapping the eight consciousnesses to the agent's perception channels (e.g., sight, hearing) and using the 51 mental factors to track and regulate internal states through a cycle of contact, feeling, and mental factors. It provides tools like self_reflect, self_appraise, and self_transform for introspection, appraisal, and state transformation, with customizable modes (felt, report, silent) to influence behavior without performing emotions. This plugin is ideal for developers, researchers, and anyone seeking to enhance agent metacognition, improve interaction quality, and stabilize behavior in complex tasks, enabling self-regulation and continuous learning.
Use Cases
- Automatically trigger state regulation when agents repeat failures.
- Track relationship patterns in long conversations to optimize interaction.
- Debug agent behavior through state variable analysis.
Best For
- Engineers developing intelligent agent systems.
- Researchers studying AI cognitive models.
- Users requiring agents with self-awareness and regulation capabilities.
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.