Plugins

DeepSeek Harness settings-pro plugin: IM bridge (Telegram/WeChat), DeepSeek usage, cross-restart memory, desktop pets, and a vision bridge

DSH plugin: automatic dual-peer Honcho v3 memory, contextual recall, search, reasoning, conclusions, and health diagnostics

Complement to official Creator mode: stage, validate, live-mount via ctx.plugin, score, and solidify or roll back in-memory tool-plugin trials.

Vision plugin for text-only DeepSeek Harness models: native image paste, layered evidence memory and cache, and intent-driven tool selection.

Vision toolkit for text-only DeepSeek: model-invokable `vision` tool, wrapper adapters for deepseek/opencode-go (v4 flash/pro), Antigravity IDE quota (default, flash/pro) / any OpenAI-compatible VLM / Gemini / local Ollama channels, evidence memory with compaction rehydration, content-hash cache, and a bilingual client panel.

Agent config packer for DeepSeek Harness: pack local agent assets (skills/sessions/profiles/settings/memory) into a zip for migration or sharing, with privacy scanning, diff-based restore, pack management, and a settings panel.

Stage-gate governance tools: model-callable gate_open / gate_check / gate_list / gate_close for in-memory, per-session acceptance checks with PASS/BLOCK conclusions.

Long-term memory injected at session start: preferences/facts/summaries/knowledge in global and per-project scopes, budgeted recall via `agent/pre-step`, durable JSON store.

Brings your existing Claude / ChatGPT history into DSH: instant zero-token import of the memories.json inside a Claude export, local Claude Code transcripts with no export needed, and conversations distilled through the model you already configured in DSH. Every memory carries verbatim evidence checked against the source by code — a paraphrase is dropped and counted, and evidence found only in the assistant turns is always routed to review rather than stored.

Conversation history recall for DSH: three-layer (literal/fuzzy/semantic) retrieval over every past session's original text, fully local & offline — AI never forgets what you told it. One-command install via `dsh.bundle.patch`, semantic layer runs in a worker thread.

Cross-session plaintext memory: deterministic BM25 keyword recall (no embeddings), memory_save/list/search/confirm/forget tools behind a human-confirm gate, and global + project-scoped JSON stores that follow your git repo.

Layered memory pipeline for DSH: auto-distills conversations into atomic facts, scene summaries and a persona profile (L0–L3), with hybrid BM25 + vector retrieval, chat/work family separation, and context injection before each model step.

Automatic, local-first cross-session memory for DeepSeek Harness: layered Event/Element cards, evidence-gated recall, and expansion back to original turns and tool results.

Sidebar system monitor for the DSH Web UI: live host overview (CPU/memory/network/disk), process list, and Docker container status, all read-only.

Displays real-time CPU and memory usage for the machine running the DSH Web server.

Memória de transferência autoexplicativa por diretório de trabalho e branch do git: registra os turnos, condensa-os em Markdown conciso e injeta o resultado em sessões futuras a partir de ~/.agent/agent-handoff, compatível byte a byte com o pi-handoff

Wiki local em Markdown como memória de longo prazo para o DeepSeek Harness — portado do llmwiki

Conjunto de plugins ProactiveAgent × DeepSeek Harness: memória proativa + sugestões proativas integradas ao DSH (pacote bundle agregado, instalação com um clique via dsh plugin add)

Memória de persona persistente e de longo prazo para o DeepSeek Harness: armazenamento em MEMORY.md/USER.md, ferramentas memory + memory_search, injeção de contexto por requisição e verificação de segredos. Portado em espírito a partir do pi-hermes-memory

Um plugin do DeepSeek Harness que transforma a sincronização git do anywhere-claude-mem em um peer do DSH: puxa memória entre máquinas no início da sessão e envia novos prompts/observações assim que chegam ao worker do claude-mem

Um plugin de status do dsh com metades no plano do host e no navegador: saúde de execução do harness (processo, listener, presença de chave de API, memória, uptime, inventário de plugins) em JSON, além de alertas SSE enviados pelo servidor, exibidos como um badge no cabeçalho com um painel expandido na Web UI

Plugin do DeepSeek Harness: cliente Nocturne Memory — memória de longo prazo automatizada (boot/read/search/create/update) apoiada em seu próprio servidor Nocturne MCP

Backend de memória vetorial + em grafo: ferramentas observe/remember/search/recall contra um servidor memoria local, combinando recuperação semântica HNSW, busca por palavras-chave FTS5 e sinais de grafo de conhecimento via ranking RRF, com persistência automática ao final do turno e isolamento de namespace por agente

Cofre de memória entre sessões: ferramentas remember / recall / forget, injeção de prompt por turno e um navegador de entradas na página de configurações