@chaoyuzhang/dsh-auto-router
Configurable Auto model routing tier for DeepSeek Harness
51 results
Configurable Auto model routing tier for DeepSeek Harness
Budget-LLM pool router for DeepSeek Harness: usage ledger, health tracking, Thompson-sampling adaptive failover across free-tier models, quality tier ladder, executor pool, paid-guard, web panel + reroute badge. 穷鬼路由器。
Tiered auto-review for DeepSeek Harness (DSH): static-rule safety net + LLM reviewer + human fallback. Auto-approve safe actions, auto-deny irreversible ones, ask a human for the rest. PURE VIBE CODING - not audited, use at your own risk.
Automatic model routing for DeepSeek Harness: a virtual `smart` model classifies every request by difficulty (hard / normal / easy) and by vision need, then delegates to the models you already configured in Settings → Models. 内置智能路由:三级难度 + 视觉自动路由,模型从「设置-模型」已配置模型中选取,虚拟 smart 模型零配置接入。
Balance & usage card plugin built on the DeepSeek Harness for its Web GUI (not affiliated with DeepSeek): account balance with color tiers, plus live per-conversation token usage and cost (peak/off-peak priced).
Aura Vision — free vision OCR plugin for DeepSeek Harness web profile: Zhipu GLM-4V-Flash (free tier), adaptive tile recognition for long documents, history with favorites and Markdown/Excel/Word/PNG export.
DeepSeek Harness relay plugin for reasoning-effort autofill, /models sync, and a capability-aware Fast toggle.
Tiered micro-clear layer for DeepSeek Harness compaction: clears whitelisted re-runnable tool results to placeholders before the official summarization tier, freeing context with no model call.
Local OpenAI-compatible model-server discovery + tiered reasoning-metadata resolution for DeepSeek Harness (dsh).
Manual budget-targeted context compression for DeepSeek Harness: the conversation model picks ranges, a cheap flash-tier route writes the checkpoint summaries.
Mention a server with @ in a DeepSeek Harness session and run commands on it over SSH, behind a two-tier approval gate that fails closed.
Commitment write-gate for AI coding agents: two-tier (deterministic + LLM judge) pre-execution policy. Engine-agnostic core with a DeepSeek Harness (dsh) adapter.
Oh My DSH — tiered model routing for DeepSeek Harness: think/build tiers, an automatic vision tier for images, and image generation through an ordinary chat model (e.g. gpt-5.6-luna on opencode-go), all configured from an 'oh my dsh' settings tab. Works in the Web (Desktop) surface and in the terminal (TUI).
BSP coding swarm plugin for DeepSeek Harness: nested 3-tier swarms (root + squad leads on deepseek-v4-pro, up to 100 concurrent workers on deepseek-v4-flash) with barrier-synchronized phases, worktree isolation, guardrail grants, budget caps, and cache-stable prefixes
DSH-Plugin: multi-engine cross-validated search — parallel queries across Brave/Tavily/DuckDuckGo/Wikipedia (Bing/Mojeek optional), URL dedupe, confidence tiers.
DeepSeek Harness (dsh) plugin that registers Lynkr — a self-hosted, multi-provider LLM tier-routing gateway — as a custom OpenAI-compatible provider in settings.yaml.
DeepSeek Harness plugin: peak/off-peak price watch, live account balance with tiered low-balance alerts, and a top-up button for the main web UI
Tiered task queue for DSH↔Hermes coordination: queue.json is the single source of truth; tasks are enqueued by DSH mechanisms (review/lesson/reset) and consumed by the Hermes-side cron with a lease/claim model. Also provides the busy-mutex for serializing Hermes calls from the approve fast path.
Two-tier image reading for text-only models in DeepSeek Harness: fast local OCR (RapidOCR, offline) first, then a vision model via modlens as fallback.
Reflect2Evolve memory plugin: layered L1/L2/L3 memory, reflection-weighted value backprop, cross-task policy induction, skill crystallization, and three-tier retrieval for OpenClaw, Hermes Agent, and DeepSeek Harness.
Evidence-gated frontier mathematics research workflow for DeepSeek Harness
Plugin lifecycle guardian for DeepSeek Harness. Read-only diagnostics (plugin_doctor) with three-tier confidence, plus installed-profile conflict scanning (plugin_profile_scan).
Nous Portal free-tier provider for dsh: Portal OAuth (device-code) agent-key lifecycle behind an OpenAI-compatible harness route
DSH web plugin: per-model thinking-intensity tiers, provider default reasoning, modalities, gateway compat switches and static custom request headers for custom llm-pi-ai providers, edited inside the official Models settings card. 模型能力:思考强度档位 / 默认思考强度 / 支持模态 / 兼容开关 / 静态自定义请求头。