Plugins

Adds a requirements-discovery gate before product selection, version choices and competitive analysis: 2–3 clarification rounds with 2–4 decision-changing questions, validation of real pain points and hard constraints, comparison of original, alternative and minimum-cost routes, and a confirmation card before research and copywriting. Confirmed reports keep the Xiaohongshu-style structure, concrete steps, audited images and source links. Ordinary lookups and explicit user skips bypass the questionnaire. Host-only static prompt sections with no config, client UI or dependencies.

Reports DSH session token usage from the local session logs as a host-side tool: totals, input, output and cache-read split, a per-day timeline, usage grouped by day or session and optional cost estimates.

Digital Study: a private WeRead study in the DSH sidebar, in a calm light-green paper palette. The shelf folds into ten categories and carries a slightly deeper paper tint of its own; highlights and thoughts page through, copy per item, and export as front-matter Markdown (one file per book or a cross-book batch) into a folder you choose; the whole study backs up and restores as a single JSON bundle. Thoughts read like handwritten notes: looser line height, the source book shown in full, long entries folding behind a show-all control, and a same-chapter highlight block for context. A five-shade reading heatmap draws only the days up to today, keeps its numbers on one line underneath, and a year-scoped KPI dashboard sits above it. Sync is incremental: a local ledger plus per-book counts mean an unchanged library refreshes in 2-3 seconds instead of a full pull, and a banner reports progress and dismisses itself after eight seconds. Search covers shelf, books, highlights and thoughts with type filters. Reads WeRead through the official agent gateway using your own API key, keeps everything in local JSON and never writes back. Requires DSH >=0.2.2-rc.1.

MailHarbor: a multi-account IMAP/SMTP mail workspace for DeepSeek Harness with local backup, AI search and compose, CalDAV tasks and reports.

Audits and evolves agent context at four scopes: a single repository, a group of related repos, a knowledge hub, or the whole machine. Discovers projects, instruction files, skills, tools, knowledge, state, evidence and credential references; reports drift, duplicate facts, broken pointers, budget overruns, unindexed knowledge, stale state and unregistered credentials; scores each scope on its own dimensions; and turns findings into reviewable migration plans that back up every file before moving it. The format contract is a versioned spec (CTX rules, markdownlint-style IDs) following open conventions — the AGENTS.md open format, llms.txt-shaped indexes, YAML front-matter note headers — now backed by published JSON Schema contracts (draft 2020-12) for the hub manifest and the plan JSON, plus an editor modeline that validates hub.yaml with zero configuration. The hub is declarative (schema 2, zones with light profiles, sections may live outside the hub root), a one-pass check command runs scan, doctor and score, and state writes are atomic and lock-protected; scoring stays honest about missing evidence — a waiver greens the gate, never the score. Ships a stdlib-only Python CLI plus two skills; credentials are referenced by path and never printed. Requires Python 3.9+ (stdlib only) and DSH 0.2.1-rc.2 or newer, below 0.3.0.

Blocks the end of a turn while temporary files from that turn are still lying around: it registers a listener on the serial agent/turn-stopping hook and, if anything is left, hands the model the file list plus two ways out — delete them, or justify each one and move it somewhere deliberate. Candidates come from two narrow temp-file sources, plus a backup-retention channel that is on by default and flags excess .bak copies of files touched this turn. The first is the .agent-tmp/ convention directory under the session cwd: whatever is left in it is reported outright, because that directory exists only for throwaway files. The second is the paths this turn actually wrote, captured through the fs/write-intent, fs/edit-intent and fs/delete-intent hooks; those are reported only when they sit under that cwd (compared after realpath normalisation), fall outside .git, node_modules, out, dist, build, target, vendor, .venv and the Python bytecode cache directory, match a temp-file pattern that requires a separator after the prefix so checkout.js and benchmark.ts are not taken for scratch files, and the file is still present — which keeps junk already sitting in the workspace from being resurrected as a complaint about this turn. One gate per turn, subagent sessions skipped, nothing deleted unless the backup mode is explicitly set to prune, and the injected wording follows the host locale instead of assuming a Chinese one.

Agent preset that hands every task to a Claude Code subagent, so the session only writes the task brief, delegates and reports the result. Requires a locally installed, logged-in Claude Code.

A fact layer for daily heartbeat turns: diffs the curated plugin market for newly published entries and the Chinese education SRC calendars (EDUSRC rounds, Beijing education SRC seasons, the standing vulnerability-hunting exercise) for new items. Reports deltas only — the first call seeds a baseline and stays quiet.

Scheduled model-news radar: collects official release announcements from LLM vendors through RSS, Hugging Face and GitHub organisation APIs and changelog pages (a 6-vendor sample list is bundled, a user-maintained list is read from ~/.dsh/model-news-radar/sources.json), deduplicates them into daily, weekly or monthly Markdown reports and opens a DSH session that writes the summary. Read-only, no logins, no third-party dependencies.

One DSH tool that runs a command on any execution surface the machine has -- WSL, MSYS2, Git Bash, PowerShell, nushell, interpreters, throwaway containers, ssh, real terminal windows and a real console -- and reports the version each one gives and the capabilities it has.

Record lecture audio in a phone browser and transcribe it locally with faster-whisper; generates summaries, mind maps and quizzes with a wrong-question gate plus a focus report, and serves the phone UI over a built-in HTTPS tunnel from any network. Transcription needs local Python, ffmpeg and faster-whisper (the setup script installs them plus a ~460 MB Whisper small model); the public HTTPS tunnel uses a Windows cloudflared build and has no access code by default, so anyone with the link can open it.

A graph of your memory store for DSH: one node per real memory, and every edge must state why the two are connected — a quoted keyword, a shared term, a project, or temporal adjacency. Pure filesystem reads with no model in the loop; unreadable sources are reported, never replaced with fake nodes.

Caps how many DeepSeek Harness conversations run at once, so extra conversations wait in a FIFO queue instead of piling up model requests. On Windows a bottom status strip shows physical memory, commit memory and disk usage; on other platforms only the queue and physical memory are reported. A plugins-page card changes the limit.

Sets token, cost and wall-clock caps on a DSH goal and blocks that goal when a cap is reached, instead of only reporting the spend. The figures come from the usage events DSH records for the session, priced per model, with a pill beside the composer that opens a per-model breakdown. The official DeepSeek tariff is built in, with peak/off-peak pricing toggleable because it depends on the route; other models are priced from configured rates, and a model with no rate has its tokens counted and its name listed rather than being guessed at.

A dedicated background-job dashboard for the DSH Web UI: merges jobs from every session, shows producer-reported or parsed progress with an honest estimate fallback, groups by status, purpose, session or kind, streams live output, and keeps a local history ledger.

A conversation view beside Chat and Trajectory that turns a DSH session into a character stage, with Codex Pet sprites, comic bubbles and a completion report listing the files and attachments the turn produced.

Reusable report automation skill with target-template and AI-generated standard-report flows

Compare Git/SVN versions across repositories and export Chinese Markdown release reports with optional approved model analysis.

Control tower tools: dispatch a task to a role agent, read its report, verify and accept or send it back. Requires the separate agent-control-tower Python CLI (tower; it is not on PyPI — install it with pip install -e from the same repo) on the DSH process's PATH plus a tower init-ed workspace; without it every tool call fails with ENOENT.

One-person-company operations SOP for DeepSeek Harness: 8-field company profile, a daily cycle from morning brief through two execution shifts to an evening report, ledger and cash-snapshot conventions, quote/contract shapes, and six red lines with a human-approval outbox.

One main session coordinates several root sessions: a /relay command and 15 peer_* tools for batched status reports, a version-line notice, task-board claims, handoff and wake, repository watch, read-only session inspection, and a web panel. Phone notifications need the DSHA device bridge.

Adds two entries to the DSH web sidebar above the settings row: a batch picker that groups sessions by workspace, filters them by a search box and moves the selected ones into a recycle bin, and a recycle bin that restores sessions one by one or all at once, or permanently deletes them. The recycle bin keeps its own list in browser storage and hides its sessions by wrapping the host session-visibility check, so it does not use the host archive; a moved session keeps its log on disk until it is permanently deleted. Sessions that are currently running are refused and reported instead of being moved.

Shows the cost of the current session's latest exchange, meaning everything since your last message, as a CNY amount to the right of the permission control in the composer, calculated from provider-reported token usage at DeepSeek's peak and off-peak prices.

MCP server panel for the DeepSeek Harness Web GUI: the sidebar lists every @deepseek-ai/dsh-mcp-client row — including rows inserted by other plugins — and derives its connection state from the loader fiber, the mcp__<server>__* tools still registered, a TCP probe of HTTP endpoints and the last connection test; it shows the row's configuration, enables or disables it by editing only that row of the patch file, reconnects once after a failed attempt, and reports unavailable servers to the agent through a system-prompt note plus a call gate that refuses their tools instead of letting each call fail at the transport.