dsh-llm-bounded-retry
Bounded all-error request retry plugin for DeepSeek Harness
294 results
Bounded all-error request retry plugin for DeepSeek Harness
Local, zero-network, fail-closed secret preflight for DeepSeek Harness model requests.
Education toolkit for DeepSeek Harness agents: lesson-plan skeletons, quiz validation, question scaffolding, analytic rubrics, grading sheets, study plans, class statistics, flashcards and readability levels
Request image optimization, intelligent retention, and observability for DeepSeek Harness (DSH)
DeepSeek Harness plugin: injects an [IMPORTANT] chain-of-thought rule block at the very top of the system prompt of every model request (English CoT starting with I'm, no 'Let me', direct verbs).
Enhanced ask_user_question composer for DeepSeek Harness: batch tabs, number-key answers, and a confirm summary on top of the stock question flow
DeepSeek Harness plugin: content-routing LLM adapter. Text-only requests go to DeepSeek; requests containing images go to a user-configurable OpenAI-compatible vision provider. Read_image works natively for text-only models.
Notes and todos as plain Markdown in one Vault folder of your choice: /ml jots into today's journal, todos come in deadline/sleep-until/anytime flavors with auto-wakeup, /ml note distills chats into a knowledge base, /ml ask answers questions from it, /ml mail reads your work inbox incrementally into a throwaway temp dir and extracts todos/reads, /ml view fuzzy-opens any file — notes stay plain files.
HTTP 请求工具:发起请求返回状态码/响应头/响应体/耗时(Node fetch)
DeepSeek Harness (dsh) plugin: route one request to a modality-capable fallback model instead of forcing the whole session onto a single model.
A vision bridge for text-only DeepSeek Harness models: registers an `analyze_image` tool that answers questions about images via ANY OpenAI-compatible vision/multimodal endpoint (SiliconFlow, DashScope, Zhipu, OpenRouter, Ollama, ...).
DeepSeek Harness host bundle that implements slashx.request.v1 and slashx.response.v1
Minimal-anchored agent preset for DeepSeek Harness: first request keeps the real Minimal tool pair, promotion unlocks the full Standard catalog with a one-shot background-jobs notice
Bidirectional bridge: chat with your DeepSeek Harness agent from Feishu (Lark) — long-connection bot, per-user sessions, approval/question forwarding
Quote-to-composer UI plugin for DeepSeek Harness Web: select conversation text in the chat and send it together with your next question
GitHub issue/PR search and view tools (gh_search / gh_view) for DeepSeek Harness, backed by the GitHub CLI (gh). Personal plugin ported from oh-my-pi's gh-search / gh-view — not official.
DSH plugin: API retry settings page - view and edit each LLM provider's request retry policy (mode, maxRetries) over the settings wire
Select any text in a DSH chat reply to get a floating Follow-up / Copy pill — one click turns the selection into a follow-up question in the composer. · 选中聊天回复中的任意文字,浮出「追问/复制」气泡,一键把选中内容变成追问填入输入框。
DSH web plugin: a slim right-edge question-navigation rail for long conversations — every real user question gets a hoverable tick on the right edge; click to jump straight to that turn, current position stays highlighted while scrolling
职场提问
涨薪谈判
DSH plugin: infinite exponential retries on LLM requests, capped at 10 minutes per wait
Tolerate redundant sandbox_permissions requests: a tool call asking to escalate to a mode that is not strictly wider than the session's standing mode runs as-is at the standing mode instead of failing with "not strictly wider". Genuine escalations still route through the approval flow.
HTTP 请求工具:发起请求返回状态码/响应头/响应体/耗时(Node fetch)