dsh-zen-proxy
DeepSeek Harness native LLM provider plugin enabling zero-config access to OpenCode Zen free models.
Install
$ dsh plugin --profile desktop add git+https://github.com/Yee-h/dsh-zen-proxy.gitPlugin 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
- Registers native DSH llm-provider routes for OpenCode Zen free models without requiring an external proxy server or custom API keys
- Injects official OpenCode client headers and pooled authentication required by upstream gateway gates
- Runs automated 10-minute availability probes with SSE payload inspection to filter out degraded or non-working models
- Partitions models into active (`opencode-free`) and geographically restricted (`opencode-free-region`) catalog routes
- Provides reasoning effort presets (light, balanced, deep) mapped directly to upstream output token budgets
- Handles transient failure grace periods and exponential backoff to tolerate temporary 429 rate limit spikes
Useful For
- Accessing OpenCode Zen free coding and completion models directly in the DeepSeek Harness desktop model picker
- Running automated coding workflows using pooled free tier LLMs without managing API keys or running background HTTP proxies
- Tuning reasoning tokens and response lengths using tailored effort presets within DeepSeek Harness
Who It Fits
- DeepSeek Harness desktop users seeking free OpenCode Zen models without API setup
- Developers looking for seamless model provider integrations in DSH desktop profiles
Documented Limitations
- Upstream free usage limits still apply, resulting in 429 Rate Limit errors when pooled capacity is saturated
- Region-restricted models (such as muse-spark) return 403 RegionError in unsupported geographic locations
- Input modalities are restricted to text only; image inputs are not supported in the initial version
- Provider reasoning effort identifiers (light, balanced, deep) are not cross-compatible with generic provider effort labels (high, medium)
- Replaces standard DSH attribution headers with OpenCode client headers to satisfy gateway requirements
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, 1 development, 3 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
- Yee-h/dsh-zen-proxy
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- e9e3b5f · 2026-10-19
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
- 3GitHub stars
- Forks
- 4
- Open issues
- 2
- Last commit
- 2026-10-08
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Multi-account failover model pool, quota aggregator, and automation suite for DeepSeek Harness.
View plugin →Free frontier LLM provider plugin for DeepSeek Harness with upstream probing and local OpenAI-compatible gateway.
View plugin →Connects local WorkBuddy and WorkBuddy AI desktop models to DeepSeek Harness with zero configuration.
View plugin →End-to-end encrypted remote access plugin bundle for DeepSeek Harness across desktop, web, and mobile.
View plugin →