Plugins

AI-for-Science research workflow skill for DeepSeek Harness, aimed at researchers and grad students using agentic AI: multi-angle literature review with per-angle files, experiment design with a caveat list, figures and paper reading; each job returns files plus a cross-verification brief for a second model. Zero deps, MIT.

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.

Ports poteto's pstack engineering playbooks to DSH, covering planning, review, verification and release workflows.

Multi-agent orchestration agent preset for DeepSeek Harness: one chief orchestrator that never implements itself — it triages every request, has a PM expert plan the work, and delegates implementation to domain experts via a merged expert roster (list and summon tools), scheduling parallel work through a dependency-DAG taskboard and a file-based message bus, with delivery gated by an independent review. Ships 11 core experts; more expert teams (Agency agents EN/ZH, classic collections) are one-click downloads from the settings page over dual GitHub/CDN channels with sha256 verification. Published on npm as dsh-expert-orchestrator (requires DSH >= 0.1.7-alpha.2).

Adds an independent settings page to update npm-stable dshmarket through the DSH plugin manager, with version verification and best-effort recovery on installation failure.

Media and vision workspace for DeepSeek Harness: image, video and speech generation, image Q&A and element locating, long-image OCR, pixel diff, HTML-screenshot verification and video summarization, with DashScope and OpenAI-compatible providers, model pools and a workbench client.

Patch-style file editing: one apply_patch tool accepting git/unified diff (default) and Codex apply_patch syntax (opt-in), applying every change all-or-nothing after a full read-only verification pass, with sandbox-aware delete and move and a dry-run mode.

Per-session execution gears for the DSH composer: an OVERDRIVE toggle turns Turbo on for the current session only, the Fast and Max gears swap the agent's planning and verification policy and may lower the request's reasoning effort, and the Host keeps the authoritative mode plus per-turn latency, step count, tool calls and whether a verification command ran.

Tool-call discipline for DeepSeek Harness with behavior prompts, failure-driven concurrency reduction and retry guidance, post-write checks with backup rollback, high-risk command gates, change summaries and verification reminders, advisory project-memory checks, configurable project-discipline reminders after compaction or at intervals, and compliance statistics.

Provides privacy, interaction, system and data protection for DSH Desktop on Windows: protects camera, microphone and screen access, guards against desktop and input interference, protects system settings, services and other processes, and safeguards selected files and directories. Uses process-level sandbox isolation and permission controls within the existing DSH workflow, with evidence from 38 public comparison scenarios published by Finogeeks. Install through the DSH Desktop plugin manager. Windows x64 only; first activation downloads and installs the matching native runtime with SHA256 verification and Windows consent, and runs an updater process.

Workflow and automation (`workflow`): task dispatch, optional Agent Teams, and tiered verification; usage and billing (`usage`): session token tracking and automatic context folding (no bill calculation); sessions and messages (`session`): persistent fold state, protected recent messages, and watermark-triggered handoff; models and account access (`model`): six dispatch pools and optional intent-based routing (no account integration); skills (`skill`): read-only MySQL/Redis checks, commit-message generation, and requirements/design docs; development and runtime (`dev`): Host/Client settings UI and tool-call safeguards. Together, these features are designed to balance output quality with token efficiency and model-use costs.

Project autopilot for DSH 0.2.1-rc.2 and 0.2.1-rc.3, with scoped authorization, independent verification, editable memories and companion chat, including automatic installation of the compatible desktop pet fork. The bundled pet is a fork of takboo/dsh-coopanion (AGPL-3.0-or-later); do not install the upstream pet alongside it.

Bundles 19 short-video production skills (1 entry point, 1 orchestration, 17 atomic) into the agent skill catalog: Chinese TTS pronunciation fixes, text-to-image and text-to-video, image panning, composition, covers, media verification, data review and four-platform publishing. Ships skills and scripts only, so a runnable pipeline still needs ffmpeg, Python 3 + Pillow, Node.js and playwright installed locally, your own Bilibili biliup credential and WeChat Channels / Douyin / Xiaohongshu login states, and the placeholder values under skills/*/scripts/ filled in with your own. An installer's own skill of the same name keeps precedence.

Skill that types the request before it searches: it splits a question into premises, definitions and analysis units, decides which report skeleton fits (verdict, explainer, causal mechanism, comparison, survey map, metric card), confirms the direction and runs a feasibility check, then emits that skeleton instead of forcing every question into a true-false verdict — in Chinese or English. Evidence gathering never trusts model memory: it searches domestic and international ecosystems in parallel, routes each point to the primary-source ladder for its domain (policy, academic, medical, financial, disaster, food safety, science, education, sports, entertainment, judicial, and more), fetches primary pages, grades sources by type, re-searches with debunking keywords, and keeps every statement bound to a full URL and citation position. Three factual judgements can be routed to a TypeSafe Jev (System One) decision model when a key is configured, falling back to the skill's deterministic rules when it is not; a model verdict can only lower a confidence tier.

Website cloning methodology skill: a real-source-first decision tree covering static sites, React/Vue/Next content sites and WebGL/Canvas-heavy fronts, mandatory verification of executable code quoted from second-hand AI analyses, and 13 zero-dependency Node scripts for CDP browser recon, asset/font harvesting, route crawling, sourcemap hunting and pixel-diff auditing.

Modular execution discipline and autonomous steward for DSH: evidence-first routing across 44 canonical skills, durable state planning, verification gates, fail-streak recovery, and interactive web dashboard.

Evidence-driven goal verification plugin for DeepSeek Harness. Locks goals via /ultragoal and verifies completion against command exit codes and generated artifacts, preventing the model from falsely reporting task completion.

Logged-in browser automation with 170+ site adapters as deterministic one-shot commands (requires the OpenCLI CLI, npm i -g @jackwener/opencli, and its BrowserBridge browser extension) — action-evidence verification (no silent failures), site knowledge & health cards, watch monitoring, trace replay, provenance reports, and a write-approval gate.

Reference verification for DeepSeek Harness (参考文献真实性检查器): the reference_checker model tool checks a paper's bibliography (.pdf/.bib/.tex/.txt or pasted text) against Crossref, OpenAlex and arXiv, reports found / partial / not_found / error per entry, and emits corrected citations in the input's style (APA, GB/T 7714, IEEE, Vancouver, MLA, Chicago, Harvard, ACS, BibTeX).

Batch pseudonymization for medical text, CSV, JSON, and XLSX in DeepSeek Harness. Deterministic identifier rules run first; an optional OpenAI-compatible model pass adds discovery and verification. Exact-span replacement preserves file structure, tokens stay stable per value, and each batch produces an audit report plus a re-identification vault. DICOM images get header-only de-identification based on PS3.15 Annex E, with pixel data left byte-identical.

Moves a plugin set between DSH profiles (web to desktop by default): diff, pre-checks, an offline script for CLI-owned profiles, and a post-restart verification. It installs nothing itself.

Math Agent Framework for DeepSeek Harness: symbolic derivation engine (CES / quadratic forms / ODE / PDE), SymPy verification, Lean 4 compiler-verified formal proofs, and QED-style multi-agent adversarial verification, exposed as native DSH tools through a TS shell bridging a Python engine. Requires cloning the math-agent-framework repo and `pip install -r requirements.txt`, with MATH_AGENT_HOME pointing at the repo root. Every result carries provenance (engine versions, seed, tolerances) and is marked untrusted data.

Claim-verification skill that checks design documents and plans against codebase facts, escalating behavioral claims to executable logic-primitive verification for state machines and data models, with before/after regression, domain constraints, and concurrency risk mining.