long-task-runnerRun long jobs; hand off new tasks with proven experience
Install via ClawdBot CLI:
clawdbot install dhzyw/long-task-runnerGrade Limited — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Calls external URL not in known-safe list
https://github.com/dhzyw/long-task-runner.gitAudited Sep 13, 2026 · audit v1.0
Generated Oct 7, 2026
A media studio uses Long Task Runner to queue Blender or After Effects renders across overnight jobs, keeping logs bounded and validating output frames before delivery. If a render fails, resume --replay-safe restarts only the safe unfinished segments instead of re-rendering the entire sequence.
A podcast or e-learning producer submits long TTS synthesis jobs that can take hours for multiple chapters. The runner tracks per-chapter progress, exposes status revisions so editors can poll without reading logs, and validates audio artifacts exist and are non-empty before handing off to the editor.
A research team schedules expensive local model training and feature-generation jobs, sharing a GPU via a common resource directory so jobs serialize correctly. The runner enforces completion checks on checkpoint files and reports failures with numbered state records for targeted debugging.
A developer uses the runner for slow local builds, packaging, and integration smoke tests on Windows with PowerShell. Launch deduplication prevents conflicting builds, and only exit-code and artifact validations mark success, guarding against false positives from progress bars or heartbeat signals.
An operations analyst submits long-running remote batch jobs and uses the runner to track local coordination state while external tasks remain non-retryable. After a completed task, the skill offers a fresh-chat handoff summary and suggests packaging the proven flow as a reusable skill for the next independent project.
Media and audio companies deploy the skill so their in-house teams can run and monitor long render or TTS jobs without custom orchestrators. Standard-library Python keeps deployment light on Windows workstations and avoids new infrastructure costs.
Freelance automation consultants offer setup and handoff packages around the runner for small studios doing batch rendering, TTS, or data processing. They customize job.json specs, wire artifact validation, and train clients on bounded status checks.
Publishers package task-specific reusable skills generated from proven Long Task Runner flows, covering scenarios like render farms, batch TTS, and data pipelines. Buyers import the skill alongside existing quality skills, gaining safe execution and recovery without rebuilding runners.
💬 Integration Tip
Wrap the runner as a subprocess invoked by your agent framework, passing an absolute job.json path and a dedicated job directory per task, then poll status --since revisions rather than tailing full logs. Ensure the host tool's waiting semantics match host-waiting.md so you never fake wake-ups or bypass communication limits.
Scored Oct 7, 2026
Real-time search engine supporting web search, vertical domain search, parallel batch search, and URL content extraction.
Manage Feishu (Lark) calendars by listing, searching, checking schedules, syncing events, and marking tasks with automated date extraction.
cad reference tool
Search, install, and create OpenClaw skills using intelligent matching across built-in, local, and GitHub skill repositories.
Use when building CLI tools, implementing argument parsing, or adding interactive prompts. Invoke for CLI design, argument parsing, interactive prompts, progress indicators, shell completions.
Publish local skills to ClawHub with customizable version, name, path, and optional changelog using command line or Python API.