Plugins

Read-only health clinic for the installed DSH plugin set: loader health, dependency integrity, version compatibility, install-script risk, duplicates and patch integrity, delivered as a model tool, a Settings dashboard and JSON reports.

Local-first structured vision for text-only agents: images go to a local OpenAI-compatible VLM and come back as JSON evidence (summary, verbatim OCR, layout regions, entities/relations, colors, explicit uncertainty), with anti-hallucination fallback and an optional paste/upload bridge; zero cloud cost, images never leave the machine.

Data analysis toolkit: CSV/TSV/JSON profiling, IQR and z-score anomaly detection, structured summarization with quality issues, and code-free filter/group/aggregate queries.

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.

Guards the persisted session-projection cache: per-row JSON degradation so one misbehaving projection unit (e.g. a third-party plugin storing a Map) can never stall session titles or the whole cache again, plus a startup self-heal that backfills missing titles.

Create/modify/inject test assistant messages and import/repair session.jsonl(.zstd) session logs into new sessions.

Capture every upstream model API payload to JSON (debug & observability).

Runtime log panel for DeepSeek Harness: live SSE log view in Settings plus bounded date/size rotating JSONL files.

MCP server manager for DeepSeek Harness: manage MCP servers from the Web settings page, stored globally in $DSH_HOME/mcp-servers.json, connected/disconnected at runtime.

A dsh status plugin with host-plane and browser halves: harness runtime health (process, listener, API-key presence, memory, uptime, plugin inventory) as JSON plus server-pushed SSE alerts, surfaced as a header badge with an expanded panel in the Web UI

Typed, guarded container control: ps/logs/inspect/exec/start/stop and compose up/down with JSON output, project-aware targeting, and approval-gated destructive ops.

Connects the Xiaozhi voice assistant to DSH Web over MCP: DSH is the tool provider, weaving 35 DSH Web endpoints into 16 voice-friendly tools for workspaces, sessions, chat, models, settings and files — outbound WebSocket to the Xiaozhi MCP access point by default (no public IP or port forwarding needed), several devices bound at once, plus a DSH settings page with live status.

Federated plugin registry aggregator for DeepSeek Harness — combine npm, GitHub, and JSON catalogs into one searchable index.

A theme gallery for DeepSeek Harness (DSH): JSON-defined skins with full-screen coverage and sidebar scenery, a 15-swatch palette-picker card, a pet-widget card (7 registry-driven pets playing by the chat input), a mood-greeting card (a scrolling greeting

Persistent long-term memory for DeepSeek Harness: a JSON store of typed entries (fact, preference, entity, rule, episodic) exposed as the memory_remember and memory_recall agent tools, with recall ranked by keyword overlap, importance and access frequency.

Media-scape theme for the DSH web UI: standing/character-art and video wallpapers (mp4/webm, local + online) with boot animation, base color schemes (single-source JSON + presets + picker), capsule host-element styling, ambient particles, background music player (covers/playlists/random), typing sound effects, and a draggable dock toolbar with a unified upload entry — wallpapers and music uploaded together (suffix-classified), mid-upload no-disk buffering, pause→.part snapshots with offset resume, filename+size dedup with "skipped" notices, and true-delete removal with list refreshes.

Persona engine for DeepSeek Harness: one config.json drives the AI's self-name (per model, per tier), how it addresses you, relationship stance, character text, per-item work contracts with individual toggles, thinking-chain language, appearance and reply tone (both opt-in and per-model overridable), and long-term memory gated by human confirmation - the AI can only append proposed candidates, a human confirm line flips them into effect, and the log is physically append-only and replayable. Confirmed entries are injected as a quoted data-not-instructions block; entries tagged with a kind can instead be routed to arbitrary files on confirmation and never reach the prompt. Three prompt sections re-evaluate from disk on every assembly, so edits apply on the next step with no restart. Persona presets are shareable JSON snapshots (imports and exports SillyTavern character cards, v2/v3 read, v2 written) and deliberately never carry memory. Zero runtime dependencies, no network, no shell; any error degrades to an empty section rather than a broken session. Ships blank - installing changes nothing until you write something.

Live progress for long-running DSH tasks: a script appends JSON lines to a per-session progress file, the Web UI draws them in a floating overlay and a right-sidebar tab that also lists background jobs which reported nothing, and `dsh-progress run -- <cmd>` wraps a command that cannot report for itself.

Deterministic session evidence reports for DeepSeek Harness: /transcript writes styled single-file HTML / Markdown / JSON reports from canonical session logs, /stats prints a terminal stats card, /archive writes per-session ZIPs

Mounts the mcptoon MCP gateway as one stdio server, so the agent discovers the tools of every MCP server you have configured through mcptoon's compressed view instead of full JSON schemas on each turn. Ships a doctor CLI that checks the mcptoon executable resolves and prints the exact command/args override when it does not.

Host-level proactive wake-ups for DeepSeek Harness: the model schedules its own alarms (once, recurring interval, or cron, with optional random jitter), cold sessions are woken on time by session resume, model-initiated follow-ups are gated by quiet hours and a daily delivery budget, silent wake turns end with proactive_reclaim and fold out of model context as tombstones, declared schedule files glob-sync JSON schedules into host alarms idempotently, min_idle_seconds defers wakes until the destination session quiets down, and a web panel (settings section plus per-session tab, SSE live refresh) manages alarms and wake history — the heartbeat engine for AI companions and roleplay characters with their own daily routines and world evolution.

Local relationship-memory workbench with a confirmation gate: every AI write — extracted memories and newly created contacts alike — lands in a pending queue until the user confirms it in the GUI; materials are archived verbatim with quote-level provenance checks on AI extraction, and confirmed memories aggregate into per-contact timelines, gift occasions and reciprocity views. Ships a native relationship agent preset for conversation capture; JSON storage by default with an optional SQLite backend, zero production dependencies.

Personal workbench panel with an overview dashboard, custom kanban boards, a note pad and a file library, persisted to local JSON by the host.

Agent 行为守护桌宠：跨 agent 监控可疑外传（加密打包直传对象存储 / DNS 外泄线索 / 敏感密钥被读），发现即告警，明确确认后按「一次一个」终止目标 agent；事件写入只追加 guard-events.jsonl，并通过 DSH host 工具 + MCP(stdio) 供其它 agent 查询