dsh-plugin-subscriptions

What it is — Wires unpaid model lanes, a localhost OpenAI-compatible gateway, named fallback chains, and an on-machine usage dashboard into DeepSeek Harness.

Who it is for — You are in DeepSeek Harness shipping code or running an agent with no API key, and you need DeepSeek V4.1 Flash or Kimi K3 on the selector—or the same models exposed on localhost for other clients. Skip it where policy allows only private or leased-line egress; the unpaid lanes talk to the OpenCode Zen and Kilo public gateways.

Watch out — Reachability is whatever your machine’s egress actually gets; geo-blocked ids land in a region-limited group, while 5xx, 429, and timeouts do not drop a model from the picker. Kilo’s free pool states prompts may be logged. Marked beta, listed for dsh 0.1.5–0.1.7-rc.2. A clean-profile sandbox install registered with the harness. It invokes the shell.

The verdict — Install it, drive day-to-day traffic through keys in the bundled provider registry, and leave the free pool at the tail of a fallback chain.

What it is — Mounts upstream-refreshed model lanes into DeepSeek Harness and exposes a local OpenAI- and Anthropic-compatible proxy.

Who it is for — You are running agent work inside DeepSeek Harness, want DeepSeek V4.1 Flash or MiMo V2.6 in the current session picker, and have no provider key. If the host already talks to a company gateway through one Base URL, a second local proxy is a duplicate hop.

Watch out — In a clean profile it installed and the harness registered it. What you can call depends on this machine's egress; region-blocked models go into a limited group, and the Kilo pool says prompts may be logged. It executes shell commands; the desktop EAC lane also needs GitHub authorization. Docs list dsh 0.1.5 through 0.1.7-rc.2, status beta.

The verdict — I would keep it as a scratch lane, not the production default — the free quota sits upstream, and this plugin cannot promise it will still be there tomorrow.

What it is — dsh-purge is a DeepSeek Harness plugin that separates prompt rewriting from sandbox policies and allows swapping prompts per model.

Who it is for — In local lab security evaluations using DeepSeek Harness, users needing to separate prompt rewriting from sandbox policies are suitable for installing it. Users not performing local configuration modifications do not need to install it.

Watch out — Installation requires providing an API key or token and will execute shell commands. Sandbox testing passed in a fresh profile with no obvious issues found.

The verdict — If conducting local lab red-team assessments on Harness security configs, I would install it because it separates prompt and sandbox layers for controlled testing.

What it is — Registers locally signed-in WorkBuddy (CN and global) as DSH providers and refreshes each model's credit multiplier from upstream.

Who it is for — You already have a WorkBuddy login on the machine and you are doing coding work in DSH, but GLM-5.3, DeepSeek-V4-Pro, or GPT-5.6 still cannot be chosen there. Skip it if the official WorkBuddy app is where that account should be used, and DSH is not the place those models should appear.

Watch out — WorkBuddy is already limiting unofficial client traffic, so the bridge can stop when upstream tightens, and long-term API availability is not promised. Calls spend your own credits; the plugin does not change account state, and it does run shell commands. A clean-profile install succeeded and was registered into that profile.

The verdict — I'd add it as a part I can delete overnight—only if those models never become a load-bearing step.

What it is — Integrate Antigravity CLI agy as staffer, supporting Gemini 4 models and orchestration modes.

Who it is for — Use agy as subagent for research and code review tasks in Claude Code. Do not use it for real-time interactive tasks because agy acts as a background worker that waits for the agy job to finish.

Watch out — It requires installing Antigravity CLI first. It will execute shell commands. Benchmark results show task time increasing from 2-8 minutes to 13-26 minutes.

The verdict — I would install it because it can reduce host token spend on research tasks, though it increases task completion time.

What it is — Registers a commandcode route to a live model catalog with reasoning-effort support.

Who it is for — Users needing Command Code LLM integration in web profiles for DeepSeek Harness are suitable to install it. If you do not need Command Code models, you do not need to install it.

Watch out — Sandbox testing passed: successfully registered in a new profile after install. It executes shell commands. Requires COMMANDCODE_API_KEY. No obvious issues found.

The verdict — I will install it when I need real-time model directory and browser login support, because it allows multi-account rotation and session cost estimates.