dsh-subscription-search
Run the following command in DeepSeek Harness:
dsh plugin install shaomingbo/dsh-subscription-search
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install shaomingbo/dsh-subscription-search in your DeepSeek Harness terminal to install the plugin (source: https://github.com/shaomingbo/dsh-subscription-search).
About this plugin
In DeepSeek Harness, web search is often fragmented across providers—Ollama, Exa, DeepSeek, ChatGPT, and Grok each operate in isolation with no shared orchestration or failover. dsh-subscription-search introduces a configurable search chain (search-chain/v1) that wires these backends into a single, policy-driven pipeline, turning ad-hoc per-provider calls into coordinated dispatch.
The chain follows a default order of ChatGPT, Grok, Ollama, Exa, and DeepSeek. Exa, DeepSeek, and Ollama are built-in adapters resolved through standard DSH credential refs; ChatGPT and Grok are optional callable adapters that an account plugin registers at runtime, and the package starts and searches fine without them. The chain owns enabled flags, per-backend and total deadlines, automatic fallback, cancellation, empty-result semantics, and structured diagnostics. The Ollama adapter calls the hosted web_search endpoint, draws on the free Ollama account quota when an API key is present, and falls through transparently when the quota is exhausted—all without requiring a restart.
This plugin targets DSH users who rely heavily on multi-source web search: you want a single request to fan out across engines, automatic failover when a backend is unavailable, a unified policy panel for ordering and timeouts, and a search component that stays out of OAuth, credential storage, and quota UI. If you only need single-engine search or already have a full orchestration layer in place, the incremental value here is limited.
Use Cases
- Automatic multi-engine web search orchestration with staged fallback
- Unified management of backend ordering, timeouts, and enable flags
- Prefer free Ollama quota before consuming Exa credits
Best For
- Power users of DSH who rely on multi-source web search
- Developers needing search failover without building orchestration from scratch
- Operators who want centralized search chain policy and diagnostics
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.