Run concurrent batch processing in tmux with JSONL output to disk.
npx skills add https://github.com/snqb/my-skills --skill longrun
Please help me install this Agent Skill. Skill: longrun Source: https://github.com/snqb/my-skills/tree/main/longrun Command: npx skills add https://github.com/snqb/my-skills --skill longrun
Batch processing and long-running tasks can stall interactive sessions and create fragmented, hard-to-audit logs. longrun provides a structured pattern to run concurrent scripts inside tmux, streaming results to disk for resume and monitoring.
Create a Python script that processes a list of items concurrently and streams results to /tmp/TASKNAME.jsonl.
High-intent search queries and answers about installing and using this skill.
Batch processing tasks like API calls or scraping are handled by running asyncio concurrency inside tmux. This streams results to disk as JSONL files, preventing session stalls and creating auditable logs for large item lists.
Yes, batch processing supports resume capability by scanning existing JSONL outputs and skipping completed items. When you restart the concurrent task, it automatically detects previously processed items and continues from where it stopped.
Monitor concurrent batch processing by streaming per-item results to a JSONL file. This output format allows you to watch for new entries in real time and use the structured logs for post-processing analysis.
Yes, this batch processing approach is designed for data transformations and tasks that take longer than ten seconds. It uses asynchronous concurrency with a configurable level of parallelism to handle large item lists efficiently.
Manage parallelism when processing large lists by using an asyncio-based semaphore that provides a configurable level of concurrency. This limits simultaneous operations, preventing resource exhaustion during large-scale batch processing.
archibateImplement asyncio patterns for concurrent I/O and task orchestration.
dathereConvert tabular data files between CSV, TSV, Excel, JSONL, and Parquet formats.
kngender5Orchestrate parallel Hermes agents with dmux and tmux session management.
bennybennisonSelect Python concurrency models for I/O-bound, CPU-bound, and parallel workloads.
