dsh-model-extended
Configure per-model reasoning-effort ranges and accepted input modalities directly in DeepSeek Harness model catalogs.
Install
$ dsh plugin --profile web add .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.
Key Capabilities
- Per-model configuration of reasoning effort ranges (off, low, medium, high, max) and default effort
- Per-model configuration of accepted input modalities (e.g., text, image)
- Explicit declaration to disable reasoning capabilities on non-reasoning models
- Pre-request runtime validation rejecting unsupported reasoning effort levels with UNSUPPORTED_REASONING_EFFORT
- Integration into both DeepSeekModelsEditor and custom provider ModelListEditor interfaces
- Reversible bundle patching mechanism with automatic backup creation and verification
Useful For
- Preventing invalid reasoning effort parameters from causing runtime request rejections on endpoints that only support specific reasoning tiers
- Customizing multimodal vs text-only capabilities per model row under unified provider endpoints
- Managing heterogeneous model suites under DeepSeek and pi-ai custom provider configurations
Who It Fits
- DeepSeek Harness users with multi-model custom provider setups
- Developers configuring diverse LLM endpoints with varying reasoning or modality capabilities
Documented Limitations
- Runtime overlay only takes effect for providers configured via settings namespaces (e.g. llm-deepseek)
- Requires patching the local DSH editor bundles on startup, which must be reverted prior to package removal
- Settings files must not be processed by strict YAML 1.1 parsers that coerce unquoted 'off' values into booleans
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 0 runtime, 0 development, 1 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- lengxiaoyu6/dsh-model-extended
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- dc9a638 · 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
- 0GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-09-16
- Last release
- 2026-09-12
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Command Code LLM provider bundle for DeepSeek Harness with model catalog, quota tracking, and multi-account rotation.
View plugin →Free frontier LLM provider plugin for DeepSeek Harness with upstream probing and local OpenAI-compatible gateway.
View plugin →Use ChatGPT, Claude, Grok, GitHub Copilot, and Antigravity subscriptions as DeepSeek Harness LLM providers via OAuth.
View plugin →Interactive reasoning-effort slider and runner visual integration for DeepSeek Harness model selectors.
View plugin →