Plugins

Integrates TypeSafe Jev for skill and file selection, task supervision, shared-finding corrections, tool-output filtering, approval assistance, and recorded-stage navigation in DeepSeek Harness; all features are disabled by default.

Cross-instance message and event handoff between DSH instances via an interconnect server.

MCP management panel for the DSH web UI: per-server status, transport and full tool input schema, with add/edit/enable/disable/remove in global and project scopes, reconnect for global servers, a five-server built-in catalog, and config import from ten other clients with a preview. A project's MCP servers share one connection per (project, serverName) — the first session opens it and the last one closes it — so concurrent sessions reuse a single mcp-client instance; the panel shows how many sessions share it and marks a config change as pending until the connection is rebuilt.

Isolated-environment browser verification skill for DSH plugin development: a throwaway DSH_HOME, its own `verify_<random>` profile, its own port and its own browser instance in one command, cleaned up on exit, so the profile you are actually using is never touched; ships a zero-dependency raw-CDP driver (snapshot / click / screenshot / eval) and an optional isolation audit.

Agent2Agent protocol v1.0 for DSH: run a standard A2A server to let other agents call this instance, connect to remote A2A agents as clients, and manage both from a visual settings page.

Expert-mode agent preset for DeepSeek Harness (v0.9.2, npm dsh-expert-mode, bilingual EN/ZH): a chief coordinator plus 17 domain-expert subagents with automatic task delegation. Features: taskboard scheduler (file-system task state machine pending/ready/running/done/failed, dependency DAG, atomic claim, retry, crash recovery), quality gates (5-stage pipeline for high-risk tasks: requirement clarity, implementation, verification, independent review, integration, with 2-round rework limit), Five-Anchor constraint (review/convergence/anti-drift/collaboration-check/resource-awareness per turn), Near-distance Guidance (identity/task/output template per expert), progressive disclosure (~28% token savings), expert persistence, inter-expert file message bus (direct P2P comm, zero coordinator relay), cross review, experience pool, fast-track for simple tasks, fault recovery with auto-retry. Experts: data analyst, copywriter, legal review, product manager, frontend, UI/UX, architect, social media ops, growth hacker, quant finance, finance, backend, DevOps, database, QA, security.

SearXNG-backed web_search provider for the web seam: free, self-hosted, key-less metasearch through your own instance's JSON API, with a bundled loopback-only docker-compose example.

Clean desktop shell for DSH: launch by double-clicking a desktop shortcut like any normal desktop app, live backend monitoring with one-click tray start/restart/stop (progress dialogs), offline auto-reconnect, single instance. Reuses your existing web profile, no restyling of the web UI. Windows + macOS.

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.

COM-driven real Microsoft Office automation for DeepSeek Harness: drives a running Excel/Word instance through OfficeMCP — VBA macros, pivot tables, live formula recalculation, and deep document layout; plus an accounting suite (double-entry journal posting with debit/credit balance check, and SUMIF general ledger).

Right-hand widget rail for DeepSeek Harness Web UI: live session stats, task/context/heatmap/quote widgets, OpenCode Go quota, multi-column grid layouts (1/2/4 columns) with 2x4 long tiles, a continuous magnification wave, and per-instance sizing; widgets are installable, reorderable and resizable from Settings > Components.

Points web_search at a self-hosted SearXNG instance instead of the native route: pages until maxResults is satisfied, and reports degraded coverage when engines fail so a thin result set is not read as a thin web.

DSH updater + official docs sync in Settings: one-click npm update of @deepseek-ai/dsh with live progress (npm output stream, elapsed timer, monotonic bar), multi-copy protection that only touches the running instance, source-tree mode via git pull, one-click PowerShell 7 install on Windows, incremental sync of the official docs/ with progress, and the dsh_docs_search / dsh_docs_read model tools.

Spec-Driven Development pipeline for Odoo: 15 tools take a request from an immutable spec through architecture to module code or live configuration, then require a persisted verdict backed by evidence from a real Odoo instance, with fail-closed gates, human-approved batches in a declared environment, Odoo's own CSV/Excel importer, and a journal that reverses the data written through the plugin.

Honcho memory plugin by Plastic Labs: injects the user profile, session summary and relevant conclusions at session start, uploads each turn to Honcho in the background with secrets redacted, and adds honcho_search / honcho_chat / honcho_remember tools plus a /honcho status command; configured through the ~/.honcho/config.json shared with the other Honcho integrations, against hosted honcho.dev or a self-hosted instance.

Web search pool for the native web_search tool: fails over across 8 free/paid backends with quota-aware circuit breaking, settings-page API key management, custom SearXNG instances, failover/rotate strategy switching and live SerpApi quota.

Bridge DeepSeek Harness agent sessions into Feishu (Lark) chats over the official SDK long connection, with a streaming reply card, many bots per instance, and mention-gated group replies.

Read-only Gitea and Forgejo tools over the REST API both projects share: instance info, repository listing, issue and pull request search and read, PR diffs and changed files, and Actions runs, jobs and job logs. Tool metadata is available in English, Traditional Chinese, Simplified Chinese and Japanese.

Auto maintenance and resilience for DeepSeek Harness: startup self-diagnosis, automatic repair of missing directories, scheduled configuration backups with rotation, one-click restore, smart-start port detection, plugin-change self-check with automatic rollback, startup failure rescue, and a health watchdog that restarts a stuck-but-alive instance. Ships as a Cordis plugin plus a standalone CLI.

Embeds an Overleaf site (overleaf.com or a self-hosted instance) into a conversation tab through a same-origin reverse proxy, with cookie or popup login, selection quoting, caret insertion, compile-log driven fixes, outline navigation, and .bib / .tex synchronization.

Terse prose for DeepSeek Harness — compress the agent's own output so every reply costs fewer tokens (layer 4). Style, never substance.

Read-only SonarQube Community Build tools: instance status, project Quality Gate for a branch or pull request, issue and Security Hotspot search, single hotspot detail, and coverage, duplication or caller-selected measures. Issue and hotspot results carry a normalized location with component key, file path, line and text range.

Read-only Grafana tools that query metrics through the data source proxy: instance health, data source listing, instant and range PromQL queries, current alert state, and provisioned alert rules. Range queries downsample to a point budget; an explicitly supplied step that would exceed it is rejected rather than silently changed, so the model never receives data at a resolution it did not ask for.