Workflow & Automation

Engineering-discipline guard: requirements grill before the first edit, red/green test-evidence gates, forked adversary review, and a delivery report with a per-dimension verification workflow.

11 shareable workflow skills (handoff, triage, to-spec, to-tickets, wayfinder, teach, wait-what, dsh-mode-routing, ask-matt, overnight-execution, full-throttle) as an installable skill pack.

Claude Code-style permission rules engine: hard/deny/ask/allow tiers with a hard tier above full access, workspace-scoped rules, wildcard path protection, and a visual staged editor; rules persist in settings.yaml.

dsh-agent-teams turns the current DeepSeek Harness session into a captain that can assemble durable sub-agents, split a goal into dependency-aware tasks, and coordinate work through direct messages.

Two-phase agent preset for DeepSeek V4 Pro: the first request sees only bash+read (Minimal-aligned bootstrap), then the full Standard tool catalog after the first tool call or reply; ships as an installer bundle plus a manual preset directory.

Adaptive deep-research orchestrator plugin for DeepSeek Harness (official workflow engine, cybernetics/information-theory design)

🕒  Need recurring or one-shot coding work to run later without relying on an old chat? 🧭  Need each unattended run to stay inside an explicit workspace and permission boundary? 🧾  Need to inspect what ran, which revision it used, and how it ended?

① The plan sounds plausible, but one sentence hides a migration risk? ② You found several independent problems, but the only choices are Approve or Reject? ③ You want every comment to stay attached to the exact text the agent must revise?

Give a task → the agent recalls past similar solutions (condition reflex), evaluates whether they fit, and produces a dynamic plan matched against its capabilities — never hard-coded combos. Every plan auto-drafts a lesson into the experience library; when the task closes, the agent updates the outcome. The more you work, the smarter the reflex.

Hook DeepSeek Harness into a Hermes pipeline: dispatch-spec template, model-tier routing, orchestrator-run quality gates, git single-writer rule, as a SKILL.md pack (bundle installable).

DeepSeek Harness plugin: Claude-Code-style lifecycle hooks for DSH. Configure shell commands that run automatically before/after model tool calls (pre-tool / post-tool) with a JSON payload on stdin; pre-tool non-zero exits block the call.

Evidence-first operating kit: six read-only tools (capability catalog, staged workflow plans, packaged skill reader, bounded local memory search, repository release audit, release checklist) and five packaged skills - plans and audits, never remote writes.

为 DeepSeek Harness Web Profile 提供定时任务：支持准点执行、只在 DeepSeek 谷时段执行的“空闲执行”，以及每天重复执行。 Scheduled tasks for the DeepSeek Harness Web Profile: run on time, run only during DeepSeek off-peak (valley) hours, or repeat daily.

A community plugin that adds a "model approval" permission mode (approve-for-me) to DeepSeek Harness: in that mode, approval/request asks are answered by a separate reviewer model instead of a human — the reviewer decides ALLOW / DENY / ESCALATE, and the request only reaches a human when the reviewer cannot decide or fails. In every other permission mode the plugin stays silent, so human approval is never front-run by the model.

TaskSwarm (蜂群) — DeepSeek Harness 上的多智能体任务编排插件：waves/lanes 并行执行、git worktree 隔离、任务包与跨模型评审、崩溃可恢复 | Multi-agent task orchestration for DeepSeek Harness: waves/lanes parallel execution, git worktree isolation, task packets, cross-model review, crash recovery

dsh-ci is a generic, provider-neutral CI/CD plugin for DeepSeek Harness (DSH). It enables agents to inspect CI runs, extract bounded failure evidence, locate source errors, and safely rerun or cancel pipelines.

dsh-routines — scheduled agents for DSH: run a prompt on a cron, get the digest where you already are (file digests, chatnode delivery, unattended-safe)

Multi-agent math problem-solving & verification framework: brainstorm → solver iteration → multi-verifier debate → Verified knowledge base, with checkpoint resume and manual/auto intervention (also ships as an agent preset).

Kimi-Code-style swarm: batch-parallel subagent dispatch with adaptive scheduling, a /swarm command, and in-chat live progress rows.

A small, privacy-focused DeepSeek Harness plugin that gives an agent durable context about the CI run it is executing in. It currently supports GitHub Actions and GitLab CI, with a generic fallback for other environments that set CI=true.

🗺️ Every fact links to its origin — official docs, upstream repos, or community repos. When in doubt, the verbatim official copy wins. ⏱️ Last verified 2026-08-15 — official docs byte-identical to upstream master (47f9438, see SNAPSHOT.md); npm tags and the dsh-plugin topic (API totalcount climbed 2668 → 2671 during the 08-15 snapshot, 1000 repos captured, see sources.md §D.2) re-checked live. Upstream HEAD unchanged (47f9438), npm @deepseek-ai/dsh unchanged (0.1.0-rc.6).

Shared multi-agent task board (create/claim/transition/query) over a Cordis service key.

dsh-auto adds an Auto Approve permission preset to the DeepSeek Harness Web UI. Each action that requires approval is reviewed by a fresh, restricted DSH child Agent before the plugin allows or denies it.

Captain is a dual-face DeepSeek Harness plugin that exposes one synthetic captain model route. GPT is the planner and independent reviewer, while DeepSeek workers make the incremental repository changes.

Second-model AI auto-review for DeepSeek Harness approval requests: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default, fully auditable from the session log (approval/asked -> autoReview/verdict -> approval/decided).

发现问题(checkup) → 修复交付(fix) → 质量复查(review) 的对抗式闭环插件：基于官方 workflow 引擎的检查/修复/复查工具集

Long-task orchestration loop: Codex plans/scores/reviews, DeepSeek executes, Kimi supplements (vision acceptance, summarization, cross-check).

DeepSeek Harness (DSH) 代码参考检索与复用调查插件：开发前检索本地代码库与 GitHub/npm 开源项目，评估复用/改造/引入依赖/自制并询问用户，支持架构级复用。Code reference & reuse-survey plugin for DSH: search local code + GitHub/npm before building, evaluate reuse vs rewrite, ask the user, architecture-level reuse.

DSH 插件：定时循环（/loop 命令 + loop 工具 + 活动状态条）。官方 bundle 插件，dsh plugin --profile web add 安装

Task board for DSH: create tasks with project and model assignment, run them manually or on cron schedules; new sessions in a project automatically pick up its todo tasks and move them to in-review when done.

Persistent multi-model agent squads for DeepSeek Harness. Give every member its own model and tool policy, save reusable teams in Settings, then select a squad beside the composer and keep chatting normally.

Rigorous open mathematics research suite: four agent skills (rigorous-open-math-research, manage-math-research-program, math-research-workflow, lean-verify) for theorem solving with adversarial audit, research program management, pipeline orchestration, and Lean 4 formalization audit; CI-verified tests and mechanical upstream sync.

Policy-driven Git sync for DSH settings and profile configuration, with secret-aware projections, conflict review, and per-line apply controls.

Research agent suite: long-horizon goal control with auditable evidence chains, cron scheduling, multi-agent expert teams, self-evolving research memory (FTS5 + RRF recall), project workspaces, and a custom workspace UI.

Declarative task-DAG orchestration for DeepSeek Harness. This optional Cordis plugin registers a model-facing orchestratetasks tool, validates a bounded directed acyclic task graph before anything starts, and executes it through the existing ctx.workflowEngine.

Conventional-commit enforcement, changelog generation, PR summaries and branch pushes through the plain git CLI.

Cross-device workspace backup, restore, and ecosystem update via the uagent-sync CLI.

Proactive scheduling: an adaptive heartbeat plus a value network (urgency, relevance, interruption cost) decides when the agent should speak unprompted.

This is not a GitHub API plugin. It is a skill pack that teaches DeepSeek Harness how to work through GitHub engineering workflows: PR triage, review feedback, CI diagnosis, and safe publish — composing whatever GitHub/Git capabilities are already available and falling back to gh/git when necessary.

Loopback-first multi-agent message relay: HMAC-authenticated broker plus dsh plugin (relay_send/recv/peers/history), zero-dependency CLI and Python clients, wire protocol v1.0.

Calendar and hierarchical task workbench: AI-assisted intake clarification, consultation, decomposition, execution with user acceptance, and review, plus reminders and per-task AI session workspaces.

UltraCode-style multi-agent orchestration: a generatable, savable, governable, observable, resumable workflow layer.

Embedded task management engine: decision-point protocol, idea capture wall, Linear-style issue store.

Specification-driven development for DeepSeek Harness, built from native skills, commands, goals, tools, and runtime context.

Before each top-level turn closes, dsh-proof spawns a read-only verifier subagent, collects its structured verdict, and steers any non-pass gaps back into the driving agent. It is the harness's missing "is the agent actually done" gate — no other plugin can substitute for it.

Persistent multi-model agent squads for DeepSeek Harness. Give every member its own model and tool policy, save reusable teams in Settings, then select a squad beside the composer and keep chatting normally.

Workflow tools for the local aflare binary: generate, validate and run local-first deterministic YAML workflow DAGs (WAL crash recovery, Saga compensation) through the local aflare binary, with 300+ templates built in.

Adds an `auto` permission preset between workspace-write and danger-full-access: a classifier grants routine sandbox escalations once, while dangerous or uncertain requests still go to a human.

Per-project scheduled prompts run as fresh headless agent sessions, with one-time, interval, and cron schedules plus durable run history.

Publish articles to CSDN / Juejin / CNBlog directly from the session, no external server.

Cron/one-shot scheduled tasks: run shell commands or fire webhooks on a schedule, with optional ServerChan/DingTalk/Feishu/webhook delivery of results.

Overleaf-style LaTeX paper workbench: annotate any text in the rendered PDF and let a dsh agent revise the sources with compile checks and git history.

Morning Star is an Agent Plugin for harness engineering workflows: a TypeScript Harness Workflow Engine (@mstar-harness/engine) enforces deterministic workflow gates, while mstar- judgment skills drive multi-agent code delivery.

Two-level model circuit breaker with failover for DeepSeek Harness. When a model (or a whole provider) starts failing repeatedly, the plugin opens a circuit and routes the next model request to a configured fallback — no core changes, installable with dsh plugin.

- General Agent (general-agent): research, file handling, planning, and personal productivity without exposing a raw shell tool by default. - Coding Pro (coding-pro): repository-first coding with filesystem, shell, planning, delegation, workflow, and verification tools.

Condition-driven wakeup for DeepSeek Harness: the agent registers a watch, goes to sleep — even closes the session — and the sentinel wakes it when the condition happens. Every subscription and every fire is a user-visible session event, and the browser dock shows what is on duty.

A standalone dsh plugin bundle porting the omp "advisor" subsystem: a per-session reviewer model that observes the primary transcript, reviews each stepped turn with an explicitly configured model (provider + model are required), and injects severity-ranked advice (nit / concern / blocker) back into the session — without polluting or recursively reviewing itself.

Claude Science-style research workbench: ReAct research-loop engine (research_* tools), versioned artifacts with provenance (artifact_* tools), and 10 science skills for genomics/pathogens/bioinformatics.