dsh-bioinf
Run the following command in DeepSeek Harness:
dsh plugin install gsh150801/dsh-bioinf
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install gsh150801/dsh-bioinf in your terminal to install this plugin; source code is available at https://github.com/gsh150801/dsh-bioinf
About this plugin
When a bioinformatics workflow depends on hundreds or even thousands of curated skills, stuffing them all into the prompt is both expensive and wasteful. dsh-bioinf transforms 986 bio-science skills into a single BM25-style semantic search: idf-weighted multi-field scoring across name, category, when-to-use, and description, with CJK bigram tokenization so both Chinese and English queries hit the right entry. Each lookup completes in under 10 ms.
Beyond search, the plugin ships a 15-category tree for domain browsing, an on-demand skill-body provider, an optional SSH remote-GPU execution channel, and routing playbooks for three common scenarios: literature deep-review, bio-data retrieval and analysis, and remote training or prediction. Every component follows one principle: show the model only what the current task actually needs, not the entire catalog.
If you are building a bioinformatics AI agent on DeepSeek Harness and your skill library has outgrown the prompt window, dsh-bioinf is the middleware that turns a flat directory into a searchable index.
Use Cases
- Deep literature review with falsifiable downstream hypotheses
- Cross-source bio-data retrieval and comparative analysis
- Remote GPU training, log polling, and batch prediction
Best For
- Developers building bioinformatics AI agents
- Harness users whose skill library outgrows the prompt window
- Teams that mix Chinese and English queries over bio tools
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
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.