DeepSeek Harness plugin

better-model-provider

Per-model capability declaration for DeepSeek Harness: reasoning-effort levels (with wire spellings) and request modalities (text/image) for OpenAI-compatible providers — one settings section, no.

Jump to install

Source facts

Repository
sanshanya/better-model-provider
Latest update
Aug 20, 2026
Category
Models & Providers
GitHub stars
6
Format
plugin
Catalog evidence
Upstream dsh.bundle evidence
Evidence path
package.json#dsh.bundle
Checked against
0.1.0-rc.8
Upstream check date
2026-08-20

This evidence comes from the upstream catalog. This site has not installed, run, or security-reviewed the plugin.

Install

Start with a prompt that asks an agent to review the GitHub repository and source. Switch to the command if you want to install it yourself.

Copy this prompt into DSH, Codex, or another agent and ask it to review the GitHub repository and source first.

Do not install or run any commands yet. Read this plugin's GitHub repository, README, and relevant source code. Then answer the questions below clearly and directly so I can decide whether it fits my needs:

1. What is this plugin, and what problem does it solve?
2. Who is it for, and what are its typical use cases?
3. How is it used after installation? Include one minimal example.
4. What known limitations or privacy, security, compatibility, or maintenance risks does it have?
5. Give a clear recommendation: recommend, conditionally recommend, or do not recommend, with reasons.

Distinguish statements documented by the repository, inferences from source code, and unknowns. If evidence is insufficient, say so explicitly. Do not guess or simply repeat the README.

GitHub: https://github.com/sanshanya/better-model-provider
Plugin: better-model-provider
Author: sanshanya

Check the source files

Read the README and other files from this plugin directory before installing.

File explorer4 files
README.mdSource · read only
README language

better-model-provider

Per-model capability editing for DeepSeek Harness: reasoning-effort levels (with wire spellings), input modalities, and token capacities — declarations on your own routes, sparse overrides on official-catalog routes.

Custom models: edit declarations. Official models: edit overrides. Provider configuration remains official.

中文

!Editing one model row

Why

Two per-model fields stayed YAML-only: reasoningEfforts and input. Until declared, the picker shows no effort control and image sessions refuse the model (... does not accept image input). This page edits them plus contextWindow / maxTokens — one row fully configures one model.

Install

dsh plugin --profile web add github:sanshanya/better-model-provider

GitHub installs print one pnpm allowBuilds key to add, then rerun add. A local link: install must npm install && npm run build first. Restart dsh web, and the Settings sidebar gains Model capabilities.

dsh plugin --profile web rm better-model-provider

Use

1. Configure the provider + API key on the official Models page first; keys and lifecycle never move here. 2. Expand a row: pick Custom under reasoning effort, check the levels (also check off, blank wire → null, if Off should exist); check image under input modalities for vision models; fill capacities in K/M spelling (380K, 1M). Apply — the picker and image admission follow immediately. 3. Official-catalog routes: Manage official models — each field writes one sparse modelOverrides leaf; official capacities stay unless overridden; reasoning wire spellings are never guessed; Reset to official defaults lifts exactly what you overrode. 4. Manage official providers (N) lists installed catalog routes; pick one and the first override you apply materializes its profile.

Dedicated-adapter apps (built-in DeepSeek / OpenAI Codex) declare capabilities in their own adapter code and never appear here.

Compatibility

Contract line @deepseek-ai/dsh-api-remotes ^0.1.0-rc.7 (verified against the published harness); surfaces outside the contract degrade silently. Development gates, live lanes, and invariants: see [CONTRIBUTING.md](CONTRIBUTING.md).

License

MIT