What is a tasks.jsonl file?

It is a standardized configuration file containing the ID, branch name, and specific prompt for each parallel AI agent to execute via the Flocks CLI.

Can I run tasks that depend on each other?

Yes, the skill can organize tasks into sequential 'rounds,' where Round 1 must be completed and merged before Round 2 begins.

What happens after the parallel agents finish their work?

Once agents complete their tasks in isolated worktrees, you use the 'flocks land' command to cherry-pick the changes back into your main target branch.

Do the individual AI agents share context?

No, each agent runs in isolation. The skill ensures prompts are self-contained with all necessary context for the agent to complete its specific task successfully.

How does Flocks handle merge conflicts?

The skill analyzes file dependencies during the planning phase to ensure that parallel tasks do not edit the same files, preventing conflicts during the landing process.

Flocks Parallel Task Planner

Name: Flocks Parallel Task Planner
Author: 6

by6

0•

생산성 및 워크플로

Decomposes complex coding objectives into independent, parallelizable tasks for multi-agent execution via the Flocks orchestrator.

Flocks Parallel Task Planner is a sophisticated workflow skill that transforms high-level architectural goals into a structured set of parallel tasks optimized for the Flocks agent orchestrator. By analyzing file dependencies and module boundaries, it creates a tasks.jsonl configuration that allows multiple AI agents to work simultaneously in isolated git worktrees without causing merge conflicts. This skill is essential for scaling development velocity on large features, refactoring projects, or comprehensive test suite expansions by leveraging parallelized Claude instances.

주요 기능

01Support for sequential task rounds for complex, multi-stage changes

02Export of standardized tasks.jsonl files compatible with the Flocks CLI

03Strategic task decomposition to minimize file-level merge conflicts

040 GitHub stars

05Generation of self-contained prompts for isolated AI agents

06Automated codebase exploration to identify module boundaries and file dependencies

사용 사례

01Orchestrating a massive expansion of a test suite across different sub-modules

02Distributing the creation of a new feature across multiple autonomous agents

03Splitting a large-scale refactor into parallel, non-conflicting code changes

주요 기능

01Support for sequential task rounds for complex, multi-stage changes

02Export of standardized tasks.jsonl files compatible with the Flocks CLI

03Strategic task decomposition to minimize file-level merge conflicts

040 GitHub stars

05Generation of self-contained prompts for isolated AI agents

06Automated codebase exploration to identify module boundaries and file dependencies

사용 사례

01Orchestrating a massive expansion of a test suite across different sub-modules

02Distributing the creation of a new feature across multiple autonomous agents

03Splitting a large-scale refactor into parallel, non-conflicting code changes