Plugin

Ticket da scrivania a scrivania tra sessioni sulla stessa macchina. La sessione indica la propria scrivania con un prefisso numerico nel titolo; il ticket nomina una scrivania e viene preso dalla prima sessione di quel gruppo che inizia a lavorare. L’indirizzo è la scrivania, quindi il ticket non arriva a una sessione solo di passaggio.

Prompt optimizer for the composer: a button in the tool row rewrites your draft through a model channel you already configured, streaming into a panel that floats above the composer without taking up layout. Your draft changes only when you click Accept — and Accept is refused while the answer is still streaming, so a half-finished reply can never be written back. It asks first when the draft carries @ references or attachments, which a whole-draft replace would flatten into plain text. No global keyboard shortcuts, and no API key ever reaches the browser: the host half makes the call.

A project-scoped memory plugin for DSH. Knowledge authoring and the plugin's own prompts are open to editing, so users can redesign the library structure and how the model is driven without changing a line of code. By default the library is built on one core idea: a project document you design yourself (a handoff by default) is the project entity; the plugin detects which project a session belongs to, binds captured knowledge to that project, and promotes what generalizes into theme and global layers. The most natural logic there is does the pre-filtering of knowledge, with controllable compression and archiving of context.

Model attribution panel for the dsh web chat — local, zero-network structural fingerprinting (template token leaks, glitch canaries, tokenizer usage-vector matching) that scores which of 29 AI vendors a session's responses most resemble, with a per-claim evidence ledger and a 12-probe active battery.

Pannello di chimica per DSH Web. Disegna strutture in Ketcher; uno script scarica la build, renderizzata localmente nel browser senza backend. RDKit calcola la molecola corrente senza chiamate al modello: formula, masse, logP, TPSA, donatori/accettori di legami H, anelli, Fsp3, centri stereogenici, controlli Lipinski/Veber, QED, avvisi PAINS/BRENK, scaffold Murcko, ricerca SMARTS, similarità, conversione, standardizzazione e pulizia batch. La scheda retrosintesi usa un backend sostituibile (Retro* per default) tramite processo JSON in/out e segnala cosa manca se non installato, senza inventare percorsi. I calcoli sono eseguiti dal pacchetto Python del repository (rdkit), da installare separatamente con pip.

Strumenti skill on-demand per DeepSeek Harness: skill_search / skill_load / skill_ref cercano prima in una libreria a livelli e caricano il contenuto solo quando serve, senza tenerla nel catalogo residente della sessione né nel contesto di ogni turno. La scheda Skills rilegge la sessione e mostra skill effettivamente caricate e relativo turno. Nessuna dipendenza; i record sono renderizzati solo lato client e non entrano nel contesto del modello.

Security-audit facts AI coding agents cannot compute, plus a verdict gate: euthyna history attributes deleted lines to commits and flags those from security fixes (--origins finds the first introducer), euthyna coverage reports which changed symbols no test ever invoked, and euthyna gate checks each finding against six gates, downgrading any without evidence to an observation.

Aggiunge Explain alla scheda di approvazione: una chiamata al modello restituisce esattamente tre righe su operazione in sospeso, file o stato di sistema letti/scritti/modificati/eseguiti e verdetto rischio/attenzione/sicuro con una frase di motivazione. Mostra anche i dettagli delle chiamate write/edit che il renderer lascia vuoti perché legge solo il campo command.

Permessi filesystem per le chiamate shell concessi per percorso e capacità, non per nome comando. Analizza la riga di comando per leggere, scrivere, creare, eliminare ed eseguire; una capacità mancante apre una scheda di approvazione anziché un rifiuto totale. Le stesse regole sono compilate nel profilo macOS Seatbelt per processo, figli e codice non visibile negli argomenti. La scheda Approvals mostra il log di audit e indica se ogni chiamata è stata gestita da una regola, dal revisore automatico o da una persona.

Skill di aggiornamento per autori di plugin DeepSeek Harness: un indice rileva la fascia peer del chiamante e lo indirizza alla closed corridor card corretta (0.1.3-alpha.1→0.1.5-rc.1, fasi A+B; 0.1.5-rc.2→0.1.6-alpha.2, fase C), con scanner dei seam senza dipendenze nel bundle skill e nella CLI npx.

Ambitazione MCP per workspace per il DeepSeek Harness — decide quali server MCP una sessione può vedere a partire dalla directory in cui è stata aperta.

Segnala l'attività delle sessioni DeepSeek Harness a un server cc-monitor self-hosted, che mostra le sessioni Claude Code e DSH come schede di stato con codice colore in un'unica dashboard web e avvisa il tuo browser e la tua app Android quando una sessione richiede attenzione.

Input vocale pronto per la Cina per il composer. Richiede un bridge Python locale (pip install dashscope websockets, esegui bridge/voice-bridge.py) — il plugin da solo non funziona. Il microfono del browser trasmette PCM a 16 kHz al bridge, che esegue Alibaba Cloud DashScope ASR (paraformer-realtime-v2); il testo intermedio riempie la bozza al cursore, arresto automatico al silenzio, invio automatico opzionale.

Un interrupt button verde accanto alla casella di invio: interrompe silenziosamente un agente in esecuzione, che poi riassume il proprio progresso e chiede nuovi requisiti, con un interrupt prompt personalizzabile.

Plugin per DeepSeek Harness (DSH): un selettore "Lingua di pensiero" in Impostazioni → Generale che indica al modello in quale lingua ragionare (chain of thought) tramite una sezione del system prompt. Zero chiamate extra, zero latenza, 12 lingue

Recursive delegation for dsh agent trees: a preset plus Host limits bound what each subagent may do, subagents work in separate _agents/ workspaces, and the parent decides which results to integrate.

Keeps a DeepSeek Harness run going on its own: continues a turn that ended abnormally, and continues a turn logged as completed while the model tool call was emitted as plain text instead, which leaves the agent silently idle. Revives a model request that stops streaming. A wait on an approval or on a question is logged and notified, never answered.

Union web search for DSH (consolidated successor of DDDMUC/dsh-free-search, which it credits): 28 engines across three tiers — keyless free (Bing/DuckDuckGo/AnySearch/SearXNG multi-instance union, plus Exa/Tavily/Keenable/Firecrawl/Parallel keyless modes), paid-with-key (Zhihu/Perplexity/Serply/DeepSeek official/You.com/Baidu/Kimi/Aliyun/Doubao/Zhipu/Bocha/Metaso), and model built-in search (OpenAI/Gemini/Claude/Grok, explicit opt-in only) — with auto smart routing, per-engine enable/disable & fallback priority, parallel top-3 merge, multi_search, failure-aware fallback, a toggle to block the system built-in search, time filtering, platform search, web_fetch and a web settings UI. Experimental free advance_fetch: whole-page GFM-Markdown or CSS-selector/fields structured extraction with no API key and no per-call billing; fetch identity configurable (custom User-Agent, Referer, per-domain cookies attached only on domain match, never leaked across cross-domain redirects). Hosts 0.1.0-rc.2 through 0.2.1-rc.2 (17-tag exact engines.dsh enum; the rc.2 search-provider takeover goes through volatile config pinning after capability probing, never version branches). Paid engines need your own API keys (optional; free engines work keyless). npm: @drscrewdriver/dsh-union-search and bare dsh-union-search.

One settings page for every MCP server on your machine. It reads the real config file of 17 other AI tools (Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, OpenCode, Trae, Trae CN, Qoder, Gemini CLI, Qwen, Zed, Roo Code, Copilot CLI, Kiro, CodeBuddy, Continue) plus DeepSeek Harness itself, groups servers by the file that declares them, probes each one over JSON-RPC to see whether it starts and which tools it exposes, and gives every server a switch that mounts or unmounts it in DeepSeek Harness without editing the other tools' own files.

Read-only assessment of whether a Dameng DM8 database can be recovered and to which point (LSN or time), across 16 recovery scenarios, with an HTML report. Needs Java 8+ and a Dameng JDBC driver jar (DmJdbcDriver8.jar) that you provide yourself.

Adds job-role prompts to the DSH composer: type - at the start of the input to pick a role, which becomes a reference chip in the draft and is then injected as a system prompt section on every turn of that conversation, so the instruction keeps applying without showing up in the transcript. Roles are added and edited on a Settings page and stored in ~/.dsh/dsh-role-prompts.md.

Predicts your next message after each completed turn and shows it as ghost text in the composer: press → to accept, or keep typing and finish it word by word with Ctrl+→, Esc to dismiss. The prediction model follows the session by default or can be picked in the settings card, which also has an on/off switch and a speed test.

Keeps quotes out of the message body: quoting any passage from an assistant turn opens an editable quote tray above the composer, the quotes are injected into the next turn as system-prompt context (attributed with source and time), and an expandable quote row under that turn shows exactly which quotes were used. Quotes persist across page reloads and are claimed per turn; the host and client halves hand-shake on version and report a compatibility verdict.

Publishes whichever model the profile has selected in chat or on the Models page as an OpenAI-compatible HTTP endpoint: /v1/models, /v1/chat/completions (streaming SSE and non-streaming), /health. A request may name any routable model; one that omits `model` gets the current chat selection, so switching model in chat switches the endpoint. Switched on, it can also lend the model the Harness tools switched on individually, run those calls through the Harness tool runtime and feed results back up to a round limit. The listener is the plugin's own, so 127.0.0.1, 0.0.0.0 and a NAT-forwarded port all behave the same with no change to the request path. Settings cover bind address, port, API key with a generator, fallback provider and model, reasoning effort, CORS origin and timeout.