parallel-batch-executor

Parallel task execution patterns using xargs and job control for significant performance gains

5 stars

Best use case

parallel-batch-executor is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Parallel task execution patterns using xargs and job control for significant performance gains

Teams using parallel-batch-executor should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/parallel-batch-executor/SKILL.md --create-dirs "https://raw.githubusercontent.com/vamseeachanta/workspace-hub/main/.agents/skills/_core/bash/parallel-batch-executor/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/parallel-batch-executor/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How parallel-batch-executor Compares

Feature / Agentparallel-batch-executorStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Parallel task execution patterns using xargs and job control for significant performance gains

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Parallel Batch Executor

## When to Use This Skill

✅ **Use when:**
- Processing multiple independent files or items
- Running the same operation across multiple repositories
- Batch operations that don't depend on each other
- Need significant performance improvements
- Operations are I/O bound rather than CPU bound

❌ **Avoid when:**
- Tasks have dependencies on each other
- Order of execution matters
- Shared resources require synchronization
- Single task that can't be parallelized

## Complete Example: Batch Task Runner

Full implementation from workspace-hub:

```bash
#!/bin/bash
# ABOUTME: Batch Task Executor
# ABOUTME: Executes tasks in parallel using the Multi-Provider Orchestrator

set -e

# ─────────────────────────────────────────────────────────────────
# Configuration
# ─────────────────────────────────────────────────────────────────

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
ORCHESTRATOR="$SCRIPT_DIR/../routing/orchestrate.sh"
PARALLEL=1
LOG_DIR="$SCRIPT_DIR/logs"

# Colors
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
CYAN='\033[0;36m'
NC='\033[0m'


*See sub-skills for full details.*

## Performance Tuning

### Optimal Parallelism

```bash
# CPU-bound tasks: match CPU cores
PARALLEL=$(nproc)

# I/O-bound tasks: 2-4x CPU cores
PARALLEL=$(($(nproc) * 2))

# Network-bound tasks: higher parallelism
PARALLEL=20

# Memory-constrained: limit based on available RAM
AVAILABLE_MB=$(free -m | awk '/^Mem:/{print $7}')
TASK_MB=100  # Estimate per task
PARALLEL=$((AVAILABLE_MB / TASK_MB))
```
### Throttling

```bash
# Add delay between task starts
printf '%s\n' "${items[@]}" | xargs -I {} -P "$PARALLEL" bash -c "
    sleep 0.1  # 100ms delay
    process_item \"{}\"
"
```

## Resources

- [GNU xargs Manual](https://www.gnu.org/software/findutils/manual/html_node/find_html/xargs-options.html)
- [GNU Parallel Tutorial](https://www.gnu.org/software/parallel/parallel_tutorial.html)
- [Bash Process Substitution](https://mywiki.wooledge.org/ProcessSubstitution)

---

## Version History

- **1.0.0** (2026-01-14): Initial release - extracted from workspace-hub batchtools

## Sub-Skills

- [1. Basic Parallel Execution with xargs (+2)](1-basic-parallel-execution-with-xargs/SKILL.md)
- [4. Progress Tracking (+2)](4-progress-tracking/SKILL.md)
- [1. Always Set a Default (+4)](1-always-set-a-default/SKILL.md)

Related Skills

xurl

5
from vamseeachanta/workspace-hub

X/Twitter via xurl CLI: post, search, DM, media, v2 API.

xitter

5
from vamseeachanta/workspace-hub

Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.

research-paper-writing

5
from vamseeachanta/workspace-hub

End-to-end pipeline for writing ML/AI research papers — from experiment design through analysis, drafting, revision, and submission. Covers NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Integrates automated experiment monitoring, statistical analysis, iterative writing, and citation verification.

cli-productivity

5
from vamseeachanta/workspace-hub

Essential CLI tools and shell productivity patterns for efficient terminal workflows

usage-tracker

5
from vamseeachanta/workspace-hub

Track and analyze usage metrics with timestamped logging, reporting, and trend detection

state-directory-manager

5
from vamseeachanta/workspace-hub

Manage persistent state directories with XDG-compliant paths and cleanup for bash scripts

json-config-loader

5
from vamseeachanta/workspace-hub

Configuration file parsing patterns for bash scripts (INI, key=value, JSON)

interactive-menu-builder

5
from vamseeachanta/workspace-hub

Build multi-level interactive CLI menus with navigation and selection for bash scripts

git-sync-manager

5
from vamseeachanta/workspace-hub

Multi-repository git synchronization and fetch-pull-push patterns for batch operations across workspaces

complexity-scorer

5
from vamseeachanta/workspace-hub

Score task complexity using keyword matching, heuristic analysis, and configurable threshold rules

bash-cli-framework

5
from vamseeachanta/workspace-hub

Universal bash CLI patterns for colors, logging, headers, and error handling

wave-based-parallel-plan-execution

5
from vamseeachanta/workspace-hub

Orchestrate phase execution by discovering dependencies, grouping into waves, spawning subagents, and collecting results with optional wave filtering