Plugins

Intercepts DSH tool results before the model reads them: dedupe cache for repeated calls, a declarative decision graph that replaces large outputs with structured digests, an injection gate, and an honest token ledger that includes a negative-return account for over-compression. Measured 1400-1600 tokens saved per hit; the cumulative ledger is currently negative and the README says why, including that it cannot skip model turns.

Control tower tools: dispatch a task to a role agent, read its report, verify and accept or send it back. Requires the separate agent-control-tower Python CLI (tower; it is not on PyPI — install it with pip install -e from the same repo) on the DSH process's PATH plus a tower init-ed workspace; without it every tool call fails with ENOENT.

Adds lasting bots with their own name, photo, persona, memory, voice and model, a Bots panel in the sidebar, and a hidden helper subagent per bot that follows the conversation, searches ahead and leaves notes the bot reads; voice uses your own ElevenLabs, OpenAI, Groq, Mistral or Deepgram key. Interface and prompts in Brazilian Portuguese.

Project spaces, group chats, task boards, memory and usage for teams of DSH bots

本插件仅供娱乐使用，非实用性插件，下载前请自行选择。DeepSeek Harness 非官方社区插件（Token-Reducer?）：模拟执行任务，生成 SIMULATED 轨迹与 0 字节产物，不调用真实模型或工具/For entertainment only. Not a practical plugin — download at your own discretion.Unofficial DeepSeek Harness plugin that simulates task execution with SIMULATED traces and zero-byte artifacts.

Unofficial DeepSeek Harness port of the Academic Research Skills plugin (CC-BY-NC-4.0, non-commercial): 5 skills, 16 /ars-* commands, 4 agent roles, write-scope guard, and deterministic Python gates.

Registers the hosted SpecStream MCP server with the harness MCP manager and keeps the API key in the credential store. The server returns live web data as structured rows, each value with its source URL, and a run starts only after the quote's maximum is approved.

An Orchestrate tab in dsh-better-sidebar and 14 orch_* tools to spawn, prompt, fork and cancel agent chats, run command, scrape, http and agent tasks on local, ssh or remote runners, show provider status cards, and serve a secret-protected machine API. Requires DSH >= 0.1.1-rc.1, Node >= 22.5, the web profile and dsh-better-sidebar.

Posts a follow-up prompt into opted-in chats that fall silent, and offers a secret-protected HTTP API that posts a message into any chat. Requires Node.js 22.5+.

Adds a system-prompt rule that makes DSH agents look for existing work before they build.

A durable RAG-enhanced long-term memory layer for LLM agents, featuring holographic memory, state-slot conflict detection, and timeline retrieval. Requires Python 3.10+ and this repository's Python dependencies: create a venv and run pip install -e ".\[dev]" in the checkout before enabling the plugin (the bundled installer is a PowerShell script, i.e. Windows).

Adds the models of your own GPU Hugging Face Space to the model selector, starts the Space when you send a message, shows its start progress, and pauses it when idle or at a time limit. Requires a Hugging Face account with a GPU Space serving an OpenAI-compatible API, a Hugging Face token stored in the DSH vault as HF_TOKEN, and a paid plan for GPU hardware; Hugging Face bills the GPU time.

Adds a right-sidebar panel that lists the skills available to the current session, searches and tags them, combines them with your own saved prompts, appends the selection to the composer in one click, and can pin skills and prompts as always-on context injected through the system prompt.

The 7-7-7 execution loop for DeepSeek Harness — plan, run, capture, close, compound, with a human gate.

Agent preset with adaptive tool presentation and environment-aware shell routing. Ports modules from yjh051108/dsh-routing-suite and xiaobright/dsh-anchored-standard (the former is listed here), so do not enable it alongside those presets. Requires DSH 0.2.2 or later; high uses PTC with on-demand tool unlocks and max uses PTC with the full available tool catalog.

Checks each claim in a finished answer against the tool output that same turn read: a contradiction whose quote is found in that output steers one correction in the same turn, and claims the output does not back are flagged in the session log and on stderr. A free scan for absolute words and unmatched numbers runs first, and a turn that ran no tool never reaches the model.

SOLIDWORKS for DeepSeek Harness: dsh reads, edits, rebuilds, renders and exports the models in a SOLIDWORKS 2016 or later, on the same Windows PC or in a Windows VM on your Mac. Read-only by default; the Windows side decides what may be written.

An x_archive tool that turns an X (Twitter) post link into a permanent Markdown snapshot via the 2100laike archive service: images are re-hosted so they never break, snapshots survive post deletion, resubmits dedup-hit and return the latest version without spending quota, and refresh forces a refetch with a PAT. Needs the third-party 2100laike service; anonymous use is limited to 8 archives per day per plugin instance and a registered PAT raises that to 100 per day.

A wb_archive tool that turns a single Weibo post link (m.weibo.cn/status/...) into a permanent Markdown snapshot via the 2100laike archive service: long posts are fetched in full, images are re-hosted so they never break, snapshots survive post deletion, and resubmits dedup-hit and return the latest version without spending quota. Needs the third-party 2100laike service; anonymous use is limited to 8 archives per day per plugin instance and a registered PAT raises that to 100 per day.

Floating toolbar on selected text in the DSH conversation: copy / follow up / collect to Feishu / deliver to the inbox Base / rewrite.

Two-way agent bridge between two DeepSeek Harness hosts: either side can ask the other agent questions and run bounded dialogues over a PSK-authenticated TCP link, with a chat page on both sides and read-only tool posture by default. Both ends need the same shared key and a nodeId configured.

Compaction integrity: directive lines (never/must/必须 style markers) extracted from the compacted range are verified against the summary — verbatim or by content-word coverage — and every constraint the session silently forgot is reported and written to a JSONL ledger; pinned constraints from config are checked on every compaction.

Turns what a design, plan or document claims into a model and decides it: UML diagrams rendered, read and reviewed; state machines, protocols and data models checked as models; migrations tested for coverage and rollback symmetry; concurrency claims mined and routed to the checker that can decide them; dependency graphs audited; reports diffed.

Deterministic prompt-injection interception: tool results from file reads, repo searches and web fetches are scanned before they enter model context; high-severity findings quarantine-wrap or block, and credential-material files quarantine on the path alone.