← DeepSeek Harness Plugins
DeepSeek Harness PluginIndexed

kimi-tide

Per-step model router plugin for DeepSeek Harness that switches models based on rules, image presence, and keywords.

Terminal & TUIDeveloper ToolsSecurity & Policy
6GitHub Stars1ForksUpdated2026-09-15

Install

$ dsh plugin --profile web add ./dsh-kimi-tide-<version>.tgz

Plugin Overview & Capabilities

AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.

source-grounded
kimi-tide (月汐) is a model routing plugin for DeepSeek Harness (DSH). It enables dynamic per-step model selection across connected providers (such as Kimi and DeepSeek) instead of locking a session to a single LLM. The plugin evaluates explicit mentions (@model), keyword rules, and multimodal image guards to direct requests to coding, vision, or cost-effective models, surfacing routing decisions and quota details via an in-harness dock panel.

Key Capabilities

  • Dynamic per-step model routing based on customizable keyword rules and presets
  • Automatic redirection to multimodal/vision-capable models when images are detected
  • Explicit model invocation via mentions (e.g., @kimi)
  • Image transcription collaborative flows to pass image context to text-only models
  • Integrated DSH dock UI displaying real-time routing rationale and quota status
  • Configurable reasoning effort levels and fallback chain handling

Useful For

  • Automatically routing programming questions to coding-specialized models while keeping chitchat on inexpensive flash models
  • Safely sending image attachments to vision-supported models without manual session switching
  • Controlling token consumption and API costs across multiple configured model providers in DSH

Who It Fits

  • DeepSeek Harness users with multiple connected model providers
  • Developers seeking automated cost optimization and specialized routing across LLMs

Documented Limitations

  • Requires DeepSeek Harness @deepseek-ai/[email protected] or newer and Node.js >= 22
  • Requires candidate models to be pre-configured in DSH Models settings with valid API keys
  • Sessions under default latching mode remain pinned to vision-capable models once images are sent unless configured to transcribe or blind-answer

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 0 runtime, 0 development, 0 peer, and 0 optional dependencies.

info
License Declared

GitHub reports the repository license as MIT.

info
Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info
Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository
tafcear/kimi-tide
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
2ed02a6 · 2026-09-17

This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.

Repository Activity

GitHub stars
6GitHub stars
Forks
1
Open issues
0
Last commit
2026-09-15
Last release
2026-09-15
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app