dsh-plugin-focus
Run the following command in DeepSeek Harness:
dsh plugin install 863683348/dsh-plugin-focus
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, run the command `dsh plugin install 863683348/dsh-plugin-focus`. Source: https://github.com/863683348/dsh-plugin-focus
About this plugin
Building complex AI agents often leads to context drift and missed constraints as conversations deepen. dsh-plugin-focus addresses this by providing a persistent "Focus Board" for DeepSeek Harness agents. This plugin records objectives, hard constraints, and key decisions in a simple text file (defaulting to .dsh/focus.md). Serving as more than a todo list, it acts as a memory anchor to prevent forgetting. Featuring automatic injection, the plugin ensures the board is re-fed into the model context at the start of every turn, updating in real-time if the content changes. For security, all file operations are strictly sandboxed, ensuring the board cannot escape the session workspace. Whether handling long-term project tracking or tasks requiring high precision, developers can easily manage agent focus through this plugin, ensuring the AI maintains a clear understanding of core objectives.
Use Cases
- Agents needing to track complex task progress over the long term
- Code generation tasks requiring strict adherence to specific constraints
- Multi-turn dialogue systems needing coherence across sessions
Best For
- DeepSeek Harness plugin developers
- Architects requiring high-precision AI applications
- Researchers managing AI session context and stability
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.