Plugin

Aree di lavoro stile Remote-SSH: trasforma un host SSH e una directory remota in un'area di lavoro DSH nativa; l'agente opera il filesystem remoto direttamente tramite 12 strumenti rw_* (SFTP/exec), con percorsi confinati all'area di lavoro e verifica di known_hosts.

Gate di verifica per gli agenti DSH: ogni criterio di accettazione deve essere supportato da prove reali degli strumenti timbrate dal server prima che il gate di completamento lasci passare un obiettivo (audit consultivo, gate di applicazione, permessi durevoli).

Pacchetto di competenze dell'agente per note virali Xiaohongshu/RED: ricerca di note di tendenza, generazione/riscrittura di note, verifica, analisi degli account autorizzati, accesso tramite QR e pubblicazione controllata.

Due diligence di aziende e offerte di lavoro basata sulle prove per DSH: registra le fonti con livelli di prova, aggiunge affermazioni delimitate dalle prove, verifica l'entità legale presso registri ufficiali e produce un report PROCEED/VERIFY/STOP con checklist di verifica e domande per il colloquio.

Plugin di dominio per lo sviluppo Rust per DeepSeek Harness (dsh): piramide di verifica fmt→clippy→build+test→miri+audit(+semver) come strumenti rivolti al modello + governance a livello di grafo (8 regole anti-Goodhart, incl. densità di #[allow] / test eliminati, blind-spot tramite Cargo.toml e pub

Plugin di dominio per lo sviluppo del kernel Linux per DeepSeek Harness (dsh): piramide di verifica L1-L4 come strumenti rivolti al modello + governance a livello di grafo (anti-Goodhart / blind-spot / patch-conflict) + contratto AGENTS.md come sezione del prompt di sistema. JS puro, nessuna fase di build.

Disciplina di sviluppo spec-driven per dsh: cinque skill keel (anchor/spec/probe/build/audit), tre strumenti di revisione e sei template di spec impongono le specifiche prima del codice, la verifica delle assunzioni e il controllo dello scope.

Strumenti HTTP di recupero anti-bot irrobustiti dallo stack di scraping in produzione di SXS (milioni di richieste al giorno): rotazione del pool UA, nuovo tentativo con backoff esponenziale, rilevamento di muri anti-bot (captcha / sfide di verifica) e rate limiting adattivo — strumenti: `sxs_fetch` / `sxs_fetch_json` / `sxs_rate_status`.

Test di accettazione browser indipendenti per i deliverable degli agenti: in input una specifica JSON, in output un verdetto da Chromium reale.

Gate di verifica a calcolo reale: semplificazione matematica, dimostrazioni logiche, validazione con regex, FOL con E-prover, verifiche di macchine a stati e riparazione del codice, basati su binari compilati in Go con eseguibili Windows precompilati

DSH GUI Toolkit — Windows desktop control tools for DeepSeek Harness (DSH): screen layout, annotated screenshot, single-window capture, click / drag / type / key / scroll, semantic verification, window management, and UI Automation structured queries. Thi

Cordis-native, evidence-driven full-stack engineering workflow: inspect-first planning, explicit verification evidence, and approval-aware gating for sensitive operations.

SMS forwarding webhook for DeepSeek Harness. Phone apps (SMS Forwarder, Tasker, or any HTTP POST) push messages to a local HTTP endpoint; the plugin stores them as JSONL and exposes sms_recent / sms_search tools so the agent can read SMS verification codes during tasks. Auto-extracts the first 4-8 digit OTP from each message, supports keyword filtering, and works over LAN or Tailscale.

TypeSafe Jev (modello decisionale System One) per DeepSeek Harness, end to end: jev_decision (choice/score/noul, parallelo, risposte tipizzate + confidenza), una scheda Impostazioni > Plugin per la chiave API e gli interruttori, viste inline dei tool che mostrano ogni decisione nella conversazione (risposta, confidenza, latenza, costo), un auto-guard (controlli deterministici + rischi/loop di Jev), una dashboard locale /jev e un benchmark di verifica online (96,3% su 27 domande etichettate, mediana ~283 ms su jev-latest, 2026-09-21). Onesto per design: solo API reale, nessun fallback mock.

Skill that fact-checks a claim against the open web rather than model memory: it splits the claim into independently checkable fact points, routes each point to the primary-source ladder for its domain (policy, academic, medical, financial, disaster and accidents, food safety, science, education, sports, entertainment, judicial, and more), searches domestic and international ecosystems in parallel, fetches primary pages, grades sources by type, re-searches with debunking keywords, and reports conclusion-first with a full URL and citation position behind every statement, keeping unverified and undecidable points separate from confirmed ones. Three factual judgements can be routed to a TypeSafe Jev (System One) decision model when a key is configured, falling back to the skill's deterministic rules when it is not; a model verdict can only lower a confidence tier.

Live Android device pane next to your DSH conversation: scrcpy-server H.264 streaming (WebCodecs, 30-60 fps), agent tap/swipe/text with auto-screenshot verification, uiautomator element tree with tap coordinates, host-managed recording + logcat debug sessions. FLAG_SECURE pages are detected automatically with a mitigation chain and a tappable text-view fallback. Emulators and USB real phones, human+agent co-operation, session-isolated, loopback-only.

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models.

Agent preset plugin for DeepSeek Harness that triages every request, delegates implementation to an expert, and schedules parallel work through a dependency-DAG taskboard and a file-based message bus before an independent review gate. Ships with only 11 core experts out of the box; more expert teams (Agency agents EN/ZH, classic collections) are one-click downloads from the settings page over dual GitHub/CDN channels with sha256 verification.

Group-chat multi-agent orchestration console for DSH: autonomous decision-making, task dispatch, verification and iterative optimization for long tasks; bundles its own backend, installs with `dsh plugin add`.

Stability audit for installed dsh plugins: static risk grading (hook surface, startup work, inject, entry, dep ranges) plus optional isolated install verification.

Provisions the kimi-webbridge bridge daemon (auto-install with SHA-256 verification, auto-start) and registers a 25-action skill that drives the user's real browser through the local bridge.

Experience validation layer for DSH: zero-token collection of tool-failure/retry/search signals, AI refinement into verified skill books (three-layer verification), looked up at task start. Ships 10 trial skills and benchmark data (complex-task success 100% vs 60% bare).

Reliable todo panel that survives restarts (official panel only re-renders on write — fixed via projection pre-warm) with three-state completion verification: evidence exists → verified, fake evidence → blocked, no evidence → unverified badge; settings toggle to fall back to official behavior.

SMS forwarding webhook for DeepSeek Harness. Phone apps (SMS Forwarder, Tasker, or any HTTP POST) push messages to a local HTTP endpoint; the plugin stores them as JSONL and exposes sms_recent / sms_search tools so the agent can read SMS verification codes during tasks. Auto-extracts the first 4-8 digit OTP from each message, supports keyword filtering, and works over LAN or Tailscale.