Plugins

Reprend automatiquement les requêtes DSH Web interrompues : envoie un « continuer » (继续) mis en file d'attente après une panne réseau, un timeout ou un crash de l'hôte, avec classification des erreurs, backoff adaptatif, texte de reprise personnalisable et notifications du navigateur.

Convertit les articles DSH en diagrammes Excalidraw de session : diagram_create, édition avec sauvegarde automatique, export PNG/SVG/.excalidraw et relecture via diagram_read. 0.6 ajoute aperçu modifiable et index pour grands arbres à trois niveaux.

Garde et moniteur de concurrence pour DeepSeek Harness : file FIFO lorsque les requêtes modèles simultanées atteignent la limite, statistiques persistées par jour avec classification par type d'erreur, et détection d'interruption pour les flux abandonnés.

Boîte à outils vision enrichie : 14 outils vision au niveau pixel (describe, ground, detect, crop, pixel-diff, OCR, long-screenshot OCR, vectorize, colors, cutout, screenshot, present, materialize, html-screenshot) pilotés par un endpoint compatible OpenAI, avec marqueurs [图片: path] propres, classification de sécurité du contenu et auto-retry sur rate-limit.

Plugin de prédiction boursière bayésienne qui fusionne MACD, RSI, moyennes mobiles, volume et autres signaux techniques en une probabilité a posteriori de hausse, avec comparaison multi-horizons, classification de l'état de tendance et métriques de risque de position pour les actions A et américaines.

Mémoire projet DSH : crée automatiquement PROJECT.md, DECISIONS.md et SESSIONS.md au premier pas de chaque session du workspace ; relit depuis le disque et ne réinjecte qu’après modification ou perte par compaction. memory_checkpoint classe les données ; rappel limité si travail non consigné. memory_read lit journal et archive ; memory_search les interroge via BM25 sans dépendance. 200 entrées max : surplus dans SESSIONS-archive.md, jamais injecté, mais consultable, recherchable et restaurable entrée par entrée. Réglages : filtre workspace/chemin, onglets lecture seule, rechargement, JSON import/export par workspace, effacement en deux étapes, modifier/supprimer le bloc AGENTS.md, rappel PROJECT.md obsolète. Sync ; création au clic. Sans dépendances, vecteurs, embeddings ni appels modèle.

Verifiable issue-to-PR delivery chain: an 11-stage pipeline (analysis, search, code understanding, hypothesis, planning, implementation, patch ledger, tests, review, failure classification, PR build) with 4 human review gates, GitHub/GitLab/CodeArts credential connections, a 5-page workbench and a draggable AI assistant panel.

Désencombre la navigation des paramètres de DSH : regroupe les entrées de plugins tiers sous des lignes de groupes repliables, avec des groupes personnalisés de style signet, un bouton de repli et une classification automatique (tags du marché → règles de nom → votre modèle d’IA).

Routage adaptatif de modèle : classification de la complexité par requête avec routage automatique du fournisseur.

Turns DSH permissions into 9 selectable presets (3 sandbox modes x 4 approval tiers), with rule plus LLM risk classification across four risk levels, password-gated approval for high-risk operations, and a JSONL audit log.

DSH agent tools for local Ollama System One decision models: classify, answer yes/no, and score against a rubric in one batched request.

Hands secrets and PII to the agent without them entering its context, with checksum-validated extraction and masked content classification.

Sorts every loaded skill into one or more user-owned workspaces in a right-sidebar tab: each workspace is its own category list plus where every skill sits, categories are created, renamed, reordered and deleted in place, and skills are dragged or picked between them. A first run can let the profile model build the first classification, or place skills into categories the user built first; clicking a row copies its /skill-name and hovering unfolds the description the skill ships.

Lightweight crash-recovery for dsh: after a restart, it detects sessions whose last turn was killed mid-flight and re-injects a continuation prompt that first demands the agent verify real-world state before acting, so side-effecting operations are not blindly repeated. It only reacts to the genuine crash marker (turn/end reason 'interrupted', synthesized on cold read by session repair), so it never misfires on user cancels, rate limits or model errors - which is why it needs no error classification, backoff or configuration UI. Unlike general-purpose auto-continue plugins, it also covers the case where a subagent was interrupted mid-run: the parent turn has already ended normally while it waits, so a naive check sees nothing to resume and the whole subagent tree dies silently; this plugin detects the interrupted child and wakes the parent instead. Single host file, zero external dependencies, seven config constants.

Grouped skill picker for the composer. A button beside the input box opens a panel that browses every installed skill by scene (research, office, data, writing, media, dev), with auto-classification for new skills, Chinese annotations, search and drill-down; picking one inserts the /skill-name gesture.

Zero-UI-overhead DeepSeek Harness plugin that automatically classifies conversation intents and names sessions with clean emoji prefixes (optimize, fix, feature, explore, etc.).

Replaces the chat model picker with a collapsible provider-grouped list (640x720 max) that shows each model name, provider name, provider ID and exact model ID, plus a manually managed uncommon list and a separate favorites list. Classification is user-driven only: no usage tracking and no auto-sorting. Search covers model name, model ID, provider name, provider ID and account notes.

Connects TypeSafe/Jev, Cloudflare/Clef, Nace/Drex, self-hosted Laya, and custom decision providers through one interface for agent and plugin classification, filtering, scoring, and image-object localization with structured results.

session_health for DeepSeek Harness — health intelligence (outcome classification, tool-health signals, prompt heuristics, context pressure, A–F grade) computed in-process from the session store, plus a standalone CLI over the same engine.

Auto-adjust model reasoning effort per task (off/low/high/max): task classification, \[effort=...] markers, in-turn error escalation, DeepSeek peak-pricing-period capping with countdown notices, and manual-selection priority.

Apple on-device Vision tools for text-only dsh models: local OCR (zh-Hans + 30 langs), image classification, face detection, document layout, and a combined describe — 100% offline, no API key, tall-screenshot slicing.

Cross-border commerce: HS codes, customs invoices, mainland reachability. HS/HTS code lookup and classification, commercial invoices, live exchange rates, IBAN and EAN/UPC checks, amount-in-English for documents, and whether your site is actually reachabl

Chemical safety intelligence through the hosted MSDS Chain MCP endpoint: 23 tools for compatibility and mixing order, GHS hazards, PPE, storage, waste disposal, exposure limits, transport classification, multi-region regulatory compliance, SDS lookup and version diffs, and signed audit reports, each answer citing the supplier SDS and revision date it is grounded in.