← DeepSeek Harness Plugins
DeepSeek Harness PluginIndexed

kimi-tide

Per-step model router and multi-agent dispatch coordinator for DeepSeek Harness.

Terminal & TUIDeveloper ToolsSecurity & Policy
7GitHub Stars1ForksUpdated2026-10-15

Install

$ dsh plugin --profile web add dsh-kimi-tide

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 an orchestration and model routing plugin for DeepSeek Harness (DSH). It transforms model selection from a static session switch into a deterministic, per-step orchestration pipeline. Using unified route tables, 5-tier decision chains, keyword rules, and team dispatch matrices, it automatically assigns each turn or subagent to the most appropriate AI model (e.g., fast models for main orchestration, vision models for image inputs, coding models for programming, and reasoning models for review) with a real-time observability panel.

Key Capabilities

  • 5-tier decision chain for turn-level routing: explicit @ overrides, caller assignment, team dispatch roles, keyword rules with specificity sorting, and sticky main drivers.
  • Role-based team dispatch mapping subagent roles to specialized models with guardrails against unassigned delegations.
  • Automatic vision fallback and transcription workflows when images are attached to conversations.
  • Multi-model peer review workflows generating structured critiques and optional revision loops.
  • Interactive DSH dock inspection panel detailing real-time routing rationale, model quotas, and token balances.

Useful For

  • Routing daily workspace operations and bash executions to economical models while offloading code synthesis to coding LLMs.
  • Automatically switching turns to vision-capable models when screenshots or UI mockups are pasted.
  • Assigning distinct backend, frontend, and QA subagents to distinct models during team workflows.
  • Enforcing automatic peer review loops by stronger reasoning models before accepting critical code changes.

Who It Fits

  • DeepSeek Harness users working with multiple connected model providers.
  • Developers seeking token cost optimization without sacrificing specialized task performance.
  • Multi-agent workflow orchestrators managing complex subagent teams.

Documented Limitations

  • Requires DeepSeek Harness (@deepseek-ai/dsh) >= 0.1.2-rc.1 and Node.js >= 22.
  • All routed models and credentials must be pre-configured in the DSH Models settings page.
  • In image lock mode, sessions can remain locked to multimodal models unless configured to transcribe or blind-respond.
  • Dispatch guardrails reject improper agent dispatches but cannot automatically reassign tasks without teammate spawning.
  • Automated review loops increase token usage and additional host execution turns.

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
f022ae1 · 2026-10-18

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
7GitHub stars
Forks
1
Open issues
3
Last commit
2026-10-15
Last release
2026-10-15
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