Plugins

A wx_archive tool that turns a WeChat Official Account article link into a permanent Markdown snapshot via the 2100laike archive service: images are re-hosted so they never break, snapshots survive article deletion, resubmits dedup-hit and return the latest version without spending quota, a content_kind field tells text articles from sticker (image-post) messages, and refresh forces a refetch with a PAT. Needs the third-party 2100laike service; anonymous use is limited to 8 archives per day per plugin instance and a registered PAT raises that to 100 per day.

Forge quality gates for DeepSeek Harness: task gates, read-before-edit, bash hazard interception and quality scoring, enforced by the local forge CLI through DSH's typed interception points. Requires the forge CLI on PATH; without it the plugin fails open and says so with a best-effort notice at session start, and task gates also need forge init in the project.

Backup, restore and machine-migration for DSH: packs skills (which live outside ~/.dsh, in ~/.agents/skills), plugin manifests, core config, memories, todos and optionally sessions into a tar.gz with a sha256 sidecar, then imports it on another machine as a three-way merge against a baseline, listing per-file differences instead of overwriting changes made there. Credentials never enter an archive (filename deny-list of 27 names plus 7 private-key suffixes plus a content gate, enforced again on the import side), every import writes a pre-restore snapshot first, and an archive can be uploaded to any WebDAV server and pulled back on the new machine. Also includes scheduled backups, a read-only viewer for individual files, and an in-panel cloud card; no npm runtime dependencies (packing uses the system tar).

Plugin security review for DeepSeek Harness: dependency advisories, contextual static evidence, a call-time gate for high-risk plugins, and reversible review-gated enforcement.

会话记忆 → OKF 知识沉淀插件:预测驱动的神经自我学习记忆系统。把会话中高价值内容按 OKF v0.1 规范沉淀为长期记忆,跨会话自动唤起。Session-to-OKF memory plugin with neuro-self-learning (predictive recall, uncertainty-driven capture, reinforcement feedback).

Project-based learning coach: a subject-agnostic teaching engine plus swappable domain packs (Unity/C# included), shipped both as a DSH skill and as an installable bundle. The skill protocol enforces attempt-first teaching and graded hints; a zero-dependency validator rejects ability claims their evidence does not support and blocks stage acceptance while blockers are open.

Team cost captain: meters every model call (subagents included), attributes each call to a member by workspace rules, prices it with the built-in DeepSeek 2026-09 catalog (peak/off-peak, CNY/USD) or user overrides, and appends it to a local JSONL ledger. Adds hot-reloadable member budgets, 80%/100% webhook alerts, optional over-budget enforcement, a team_cost_report tool with CSV export, and a settings-page team dashboard served over a Typert RPC service.

Draggable floating ball that shows the current DSH session todo checklist live — done/total on its face (capsule mode also shows the running task title) plus a matching progress ring. Click to expand or fold the full list, right-click to switch capsule mode, and state colours mark in-progress (pulsing), done, pending and empty. Replaced plans are archived as history rows that can be hidden one by one, cleared and restored instead of counting towards progress, and other sessions can be pinned side by side. Data comes from the runtime sessions projection first with the official todo panel as a fallback, so the full list survives even when the official strip is collapsed. Text colour can be forced to auto, black or white, the ball snaps back to the bottom-right corner on demand, and it stays draggable right up to the top edge of the window. 130 shared skins — including two official dark and light skins that follow the client light/dark theme automatically, plus a per-ball sync switch so each ball either follows the shared skin or keeps its own — with Shadow DOM style isolation and position remembered per profile, identical on desktop and web.

Branch-role guard for AI coding agents: configurable integration/preview/production/archive roles, blocking direct push, force-push and delete on protected branches and agent merges into production/archive.

Read-only Odoo tools over JSON-RPC: server info, model field introspection, and a restricted search_read limited to an allow list of models whose domain field names may not contain dots, so related-record traversal is not possible. A draft-create tool is registered only when allowWrite is enabled, and is limited to sale.order and project.task with the draft state enforced by the plugin.

Safe, audited SQLite/PostgreSQL/MySQL access for dsh agents: schema introspection, enforced read-only queries, a write approval gate, and a durable JSONL SQL audit trail.

Verification gate for DSH agents: every acceptance criterion must be backed by server-stamped real tool evidence before the completion gate lets a goal through (advisory audit, enforce gate, durable permits).

Task-scoped git worktrees for DSH: durable per-repo manifest, worktree mode injects a creation instruction with your first message under an automatic worktree/ branch prefix, header badge instead of workspace clutter, and bring-back or force-remove cleanup.

Spec-driven development discipline for dsh: five keel skills (anchor/spec/probe/build/audit), three review tools and six spec templates enforce specs-before-code, assumption verification and scope control.

Memory protocol enforcement plugin: bridges the opencode-memory MCP server, forces memory_weave before tool calls, auto-ingests turns, and injects memory context into agent steps.

Neuron-inspired memory for dsh: turns buffer into a hippocampus, a nightly patrol classifies fragments into pools, and each turn injects a minimal auto-compacted handoff, the live mood (via dsh-feiyu-mood), and only relevance-filtered memories (Ebbinghaus half-life decay; recalled memories are reinforced like spaced repetition). Hard caps bound storage; past-session original text is searchable over multi-frame zstd; dsh-fish-memory stores can be imported. Optional vector retrieval (off by default, keyword fallback). The 3D panel offers four themes and pulses a signal from the center to each recalled memory.

Native generation-time Mixture-of-Agents: every turn, reference models run the same session context in parallel as pre-run drafts and one aggregator executes with full tooling; config-as-model recipes, F mirror / R trimmed views, enforced subagent gating (dual-shape session indexing), derived per-recipe vision (vision refs relay image reads to text-only aggregators), tiered cadence with separate steering and patrol rosters plus frozen reuse between beats, degeneration guard dropping degenerate reference output, anonymized model face (failure notices carry no provider identity).

Adds handoff, context-decision and context-diagnose skills, self-contained context_pressure/context_history tools (peak tokens, compaction stats), per-route realWindows with windowSource evidence, routeKey, handoffAt forced-handoff line, and slash commands.

Adds a Delete Session row to every sidebar session '...' menu (the official harness has archive/pin/rename/fork only, no durable delete). One-click Force delete ejects the session from the host live catalog, stops its background work, then removes the durable log, projection cache and registry bookkeeping - no ghost rows on any connected desktop web UI. Workspace files the session created are kept; conversation logs are permanently deleted. Third-party cloud mirrors (e.g. phone apps via Agents Anywhere) hold their own copies. Bilingual zh/en UI, trust-fenced local route. Requires DSH >=0.2.1-rc.1.

Native multi-platform accounts and model routing with staggered check-ins, event notifications, and quota forecasts. Its background worker needs a standalone Node >=22.23.3 with node:sqlite (enforced by engines.node); Qwen Office coding additionally needs Python with cryptography and wasmtime.

DSH 会话管理插件：删除（回收站恢复/彻底清除/一键清空）、强制删除异常会话、恢复归档会话、统计、继续/暂停、打开日志目录、未读标记、工作区分组与排序、上下文压缩阈值。DSH session manager: recoverable trash, empty trash, force delete, stats, continue/pause, log folder, workspace grouping, compaction threshold.

An easter-egg brake-pedal self-test. A draggable pill in the DSH web UI runs six full-force braking stomps; the bracket snaps at a pseudo-random stomp between the second and the sixth, or the run passes entirely, and then a fictional notice appears. The pedal force is derived and labelled as such, while the session figures it derives from are read from the real session log. It never intercepts, delays or takes over the real stop action.

Enforces a check-fix-recheck loop when a task is wrapping up: it auto-runs your tests/lint/audit, blocks any round that still has unfixed problems, and only calls the task done after three consecutive clean full checks, then notifies the user.

Pick the model your subagents run on — with a reasoning-effort ceiling and an output-token cap — enforced in code on every child DSH starts, whichever tool starts it.