Pending human review — the facts below come from the source repository.
Pending human review — the facts below come from the source repository.
Users who want this capability in DSH; check the source and docs before installing.
dsh plugin --profile web add github:yangyongzhen/dsh-scheduler No supported platforms were declared.
sh dsh plugin --profile <name> add dsh-scheduler # 或从本地 checkout 安装 cd ~/dsh-scheduler && dsh plugin --profile <name> add . ## 配置 在 profile 的 cordis.patch.yml（或 --patch overlay）中： yaml - id: scheduler config: enabled: true jobs: - id: daily-report # 每天 09:00 跑一次脚本，结果发 ServerChan cron: '0 9 * * *' timezone: 'Asia/Shanghai' task: type: command command: 'bash report.sh' channel: type: serverchan url: 'https://sctapi.ftqq.com/<SENDKEY>.send' - id: heartbeat # 每 5 分钟 POST 一次心跳 cron: '*/5 * * * *' task: type: webhook url: 'https://example.com/hook' payload: { text: 'dsh alive' } - id: deploy-window # 一次性任务：到点执行 at: '2026-08-16T09:00:00+08:00' task: type: webhook url: 'https://example.com/hook' ### 字段说明 | 字段 | 说明 | | --- | --- | | jobs[].cron | 5/6 位 cron 表达式（可含秒字段）；与 at 二选一 | | jobs[].at | ISO 8601 时间戳，一次性任务；与 cron 二选一 | | jobs[].timezone | 可选，IANA 时区名（如 Asia/Shanghai），仅 cron 生效 | | jobs[].task.type | command（执行 shell 命令）或 webhook（POST JSON） | | jobs[].task.command | 要执行的命令；cwd 可选工作目录 | | jobs[].task.channel | 可选，命令结果投递通道：serverchan / dingtalk / feishu / generic | | jobs[].task.url / payload | webhook 端点与 JSON 载荷（自动附加 job 与 at 字段） | ## 使用 插件提供模型可见工具 scheduler_status：列出所有已配置任务、下一次执行时间、上次执行状态与结果。Agent 在对话中询问「有什么定时任务」即可自动调用；任务本身按调度自动触发，无需人工干预。 - cron 任务错过执行窗口（如系统休眠）不会回补刷屏：触发后直接推进到下一次未来时刻。 - 命令执行有 120s 超时与 1 MiB 输出上限；投递失败只记录状态，不影响插件运行。 - 单次执行有重入保护，超长命令不会造成同任务并发。 ## 开发 sh pnpm install && pnpm build node --import tsx test/scheduler.test.mjs # 引擎单测（假时钟 + 本地 webhook sink） ## License MITNot by stars — picked by hand, one at a time.
A visual settings dashboard for DeepSeek Harness: every registered settings namespace — including third-party plugin ones the official UI never covers — rendered as editable, schema-driven forms. Early, but the right direction: stop hand-editing YAML.
A DIY trajectory visualizer for DeepSeek Harness: watch the agent’s plan and execution unfold as a graph instead of scrolling raw logs. Early, but it fills a real gap — understanding at a glance what the agent is doing.
Dispatch DSH work from Claude Code / Codex: spin up in-host DSH agent sessions with per-tier presets, watch native subagent progress, and borrow a multimodal bridge that lends the text-only harness vision and image generation. An orchestration plugin that turns DSH into a backend for other coding agents.