dsh-cron

📦 @goodandready/dsh-cron

Automated Cron Scheduling, Background Automation & Agent Execution Engine for DeepSeek Harness

🇬🇧 English •

🇷🇺 Русский •

🇨🇳 中文说明

⭐ If you like this plugin, please star it on GitHub — it shows me that the plugin is useful to you and motivates me to keep developing it.

🐛 If you find a bug or would like to request a feature, open a GitHub issue in any language — I will review your proposal and implement useful suggestions in a future plugin version.

---

⚡ Overview & The Problem

Autonomous AI agents often need to perform recurring duties: generating daily morning digests, triaging bug trackers, checking API health, syncing databases, or running periodic Git hygiene. Without a dedicated scheduler inside the harness, users must rely on external crontab wrappers, complex webhook setups, or manual intervention.

@goodandready/dsh-cron is a native full-stack scheduling and background automation plugin for DeepSeek Harness. It bridges standard cron expressions and natural interval syntax with autonomous agent execution, providing:

1. Rich Visual Task Manager — a sidebar button with a collapsible list of active jobs (next run or live state, capped and persisted), plus a full panel to inspect, filter by type/model/channel, pause, trigger, duplicate, export/import and create tasks.

2. Interactive "Create with DSH" Workflow — chat with your agent to translate high-level requirements into a well-formed scheduled task.

3. Autonomous AI Tool Calling — single unified cron tool (action: 'create' | 'list' | 'get' | 'update' | 'pause' | 'resume' | 'run' | 'delete') lets agents inspect, trigger, and manage background automation without schema bloat.

4. Robust Scheduler & Atomic Storage — built on croner with interval aliases, one-shot delays, atomic file persistence, run histories, and cost tracking.

5. Six Execution Runtimes — shell, Node.js, Python, HTTP/webhook, remote SSH and Docker, plus per-task environment variables, workspace binding and isolated git worktrees for code-modifying agent tasks.

6. Multi-Channel Delivery With Templates — one run fans out to Telegram, dsh-kanban, Discord, Slack, ntfy, Bark, PushPlus, voice (dsh-tts) and Gitea, with {variable} message templates and secrets referenced by DSH credential name.

---

🏗️ Architecture

graph TD
    subgraph Client ["Web Client Surface (DSH UI)"]
        SidebarBtn["Sidebar Clock Action(DSH Client UI Slot)"]
        Overlay["Visual Task Manager Panel(Tabs: All, Active, Paused, Completed)"]
        CreateWithDSH["'Create with DSH' Dialog(natural language task)"]
        ManualForm["Manual Task Form(Cron Expression, Timeout, Overlap, Model)"]
        SettingsCard["Settings Card(Channels, Templates, Credentials)"]
    end

    subgraph Server ["Server Runtime (Cordis & DSH Services)"]
        HttpRoutes["HTTP REST API(/dsh-cron/*)"]
        AgentTools["AI Tool Calling Gateway(cron)"]
        Scheduler["TaskScheduler Engine(Croner instances + one-shot timers)"]
        Store["Atomic TaskStore(tasks.json with atomic write)"]
        AgentRunner["Agent Session Dispatcher(Executes prompt with chosen model)"]
        Runtimes["Execution Runtimes(shell, node, python, http, ssh, docker)"]
        Notify["Delivery Router(templates + 9 channels)"]
        Secrets["Credential References(DSH credentials / ENV)"]
    end

    SidebarBtn --> Overlay
    Overlay --> CreateWithDSH
    Overlay --> ManualForm
    SettingsCard --> HttpRoutes
    CreateWithDSH -->|POST /chat/start| HttpRoutes
    ManualForm -->|POST /tasks| HttpRoutes
    HttpRoutes --> Scheduler
    AgentTools --> Scheduler
    Scheduler --> Store
    Scheduler -->|Trigger on interval/one-shot| AgentRunner
    Scheduler --> Notify

---

✨ Features & Capabilities

🐝 Swarm Delegation & Parallel Multi-Agent Execution (#350)

• Controlled Worker Pool: Configure concurrencyLimit (1..10, default 3) to execute independent subtasks in parallel without overloading system resources.
• Fault-Tolerant Decomposition: Subtask failures are isolated; the swarm aggregates all successful results and flags partial errors gracefully.
• Automated Synthesis: Aggregates worker outputs into a structured Markdown digest or delegates synthesis to a designated synthesizer model.
• Swarm Metrics & History: Run history records total, succeeded, failed counts, and duration alongside detailed per-worker execution logs.

🔍 Git Diff & Interactive Patch Review (#351, #379)

• Scoped Diff Capture: Tasks running in Git worktrees capture modified files, stat summary, and unified diffs (up to 256 KiB) automatically. Non-worktree tasks ignore repository diffs by default, avoiding storage bloat, but can opt in explicitly via captureGitDiff: true.
• Lightweight Archive Storage: Historical run archives automatically strip heavy raw diff payloads into concise excerpts with byte-budget caps (max 15MB) and automatic cleanup of orphaned .tmp.* files.
• 1-Click Discard: Revert uncommitted working directory changes safely with a single click (git reset --hard HEAD && git clean -fd).
• 1-Click Apply: Re-apply previously captured diffs directly to the repository (git apply --whitespace=fix).
• Interactive Run Preview: Color-coded diff inspector embedded directly into the execution history drawer.

🔀 Decision Nodes & Conditional IF/ELSE Branching (#352)

• Conditional Routing: Tasks can define branchRules evaluating execution outcomes (exitCode, output, output.json., status) with operators (equals, notequals, contains, gt, gte, lt, lte, regex, exists).
• Fallback Else: Optional elseTask triggers when no branch condition matches.
• Deadlock-Free Synchronization: Untriggered branch tasks are safely recorded with status: 'skipped', unblocking downstream join barriers (`joinPolicy: 'all_succes