Plugins

Automated unit-test generation: a /testgen command and generate_tests tool that scaffold, run, and fix tests until they pass (LLM + offline template generators; vitest/jest/mocha/node:test).

模型回合结束后用 LLM 生成 3 条批判性跟进建议，点击即发送 | After a model turn, an LLM generates 3 critical follow-up suggestions shown as click-to-send chips

Unofficial DeepSeek Harness LLM provider plugin for Command Code: live model catalog, reasoning-effort support, Models-page card. Ported from pi-commandcode-provider (MIT).

Automatic provider/model fallback chains for DeepSeek Harness agents when LLM requests keep failing (retry exhausted, auth, quota, rate limit)

Role-based LLM retry & fallback strategies.

Task planning with experience muscle-memory: condition-reflex recall of past solutions, LLM capability matching, and auto-persisted lessons.

Bounded, layered, approval-gated, auditable cross-session memory: a typed `ctx.memory` seam with a zero-dependency SQLite provider, a `memory` tool, and frozen snapshot injection; every write passes the approval gate and stays reconstructable from the session log.

Zero-touch runtime telemetry for DSH: every session auto-writes task, model tier, tools, failures, duration, status into a human-readable journal with a stats section (pure observer — no tools, no LLM calls, no injection).

DeepSeek Harness (DSH) plugin: a "Thinking Language" selector under Settings → General that tells the model which language to reason in (chain of thought) via a system-prompt section. Zero extra calls, zero latency, 12 languages.

Command Code Go provider adapter for dsh: scan the Go-plan model catalog dynamically and serve /alpha/generate through the harness LLM seam

Draggable floating card showing the balance/quota of your most recently used providers (DeepSeek/Moonshot/Kimi For Coding): auto-discovery, color-coded tiers, live refresh.

Learning assistant extension for DeepSeek Harness (dsh): DeepTutor tutoring for your agent — deep explanations, self-test questions, learning paths, personal knowledge-base search (RAG), and note archiving (HTTP/WS first, CLI fallback; auto-adapts local/r

Langfuse observability for DeepSeek Harness: exports session logs as OpenTelemetry traces (GenAI semantic conventions) to Langfuse's OTLP endpoint

Deterministic keyword routing, allowlisted model switching, and an isolated image_gen channel for DeepSeek Harness.

Reuses the Codex CLI ChatGPT login as an `openai-codex` LLM route and adds GPT Auth controls to DSH Web settings.

Adaptive model routing: per-request complexity classification with automatic provider routing.

Model-based permission approval: an approval-request answerer backed by a separate reviewer model.

Native LLM-provider vision bridge: images pasted in the chat are described by a vision model (Qwen3-VL via pi-ai/llama.cpp) and the text description is fed to text-only DeepSeek for the reply — image admission, routing and compaction all run through harness-native mechanisms, with an LRU description cache and 503 retry.

One-click prompt enhancement: an independent LLM call rewrites your rough draft in the composer, fully undoable.

Asynchronous, LLM-summarized context relay between DSH Sessions and authorized Agent Team Rooms, with instant snapshots and scheduled digests.

Two-tier model router for DeepSeek Harness — LLM-Judge routing, multi-model fallback chains, exponential-backoff failover, and task-level orchestration (DSH adaptation of pi-shift-router)

Bridge the vercel-labs/skills ecosystem into dsh: LLM-driven skill search, install, and lifecycle for temp/project/global scopes

Use ChatGPT (Codex), Claude, and Grok (X Premium) subscriptions as DeepSeek Harness LLM providers, with OAuth login from the web Settings page

DeepSeek Harness plugin: query AI Monitor balance / usage from local SQLite — cost-aware agent, auto bookkeeping with the AI Monitor proxy.