Skip to content
SkillModel-led1,340 tokens in use

dispatching-parallel-agents

Splits two or more independent problems across focused subagents, each with its own scope, constraints and expected output.

322,762 stars on the repository MIT Licenseobra/superpowers · skills/dispatching-parallel-agents
/plugin install superpowers@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing dispatching-parallel-agents costs

26tokens idle
1,340tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure17/25
  • Examples15/15
  • Output15/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity56
  • Maintenance68
  • Community64
  • Maturity36
  • Momentum40

Dispatching Parallel Agents guides your coding agent when it faces several unrelated failures, such as test files that break for different reasons. It groups the problems by domain, writes one self-contained prompt per domain, and sends every dispatch in the same response so the agents run at once. When they return, it reads each summary, checks the fixes for conflicts and runs the full suite.

What You Get

  • A decision flow for when parallel dispatch fits and when it does not
  • A prompt structure with a focused scope, a clear goal, constraints and an expected output
  • A worked example with six failures across three test files
  • A review step: read the summaries, check for conflicts, run the full suite

Common Use Cases

  • Several test files failing for different root causes after a refactor
  • Independent subsystems broken at the same time
  • Parallel investigation of unrelated bugs

What Makes It Unique

Many agent skills describe delegation in general terms. This one is narrow: it states exactly when to hold back from parallel work (related failures, shared state, exploratory debugging) and gives a prompt template with the usual mistakes marked, so each subagent gets everything it needs with a clean context.

Under The Hood

A single SKILL.md of instructions with a decision diagram and prompt templates. It runs no scripts of its own and reaches no outside services. The agent follows it by issuing several subagent calls in one response and then running the project’s test suite to integrate the results.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers