dispatching-parallel-agents

Use when 2 or more substantial tasks are independent, have disjoint state, and parallel execution will save meaningful time.

Skill by tsergeytovarov

IndexedMIT★ 15

Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/tsergeytovarov-dispatching-parallel-agents

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Use when 2 or more substantial tasks are independent, have disjoint state, and parallel execution will save meaningful time. Do NOT use for coupled or sequential work.

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.

Tools: No tools found
Secrets: None asked for
Install scripts: None found

Network hosts it talks to 1

codeload.github.com · found in install command

Files it touches

~/.claude/skills/dispatching-parallel-agents

What's inside

Skill name: dispatching-parallel-agents
Description: Use when 2 or more substantial tasks are independent, have disjoint state, and parallel execution will save meaningful time. Do NOT use for coupled or sequential work.
License: MIT
Instructions: 34 lines in SKILL.md · view on GitHub

Files (2)

CHANGELOG.md
SKILL.md

SKILL.md (first lines)

# Dispatching Parallel Agents

Parallelism is useful only when coordination costs and merge risk stay below the
time saved.

## Preconditions

Dispatch only when:

- at least 2 tasks have independent outcomes;
- agents will not edit the same files or mutable external state;
- each task can be briefed without the full conversation;
- useful coordination or implementation work remains for the current agent.

Do not delegate tiny tasks, related failures, exploratory debugging, or work
that depends on sequential discoveries.

## Workflow

1. Define one bounded domain per agent.
2. Provide goal, exact scope, constraints, known context, verification, and
   expected return.
3. Dispatch independent tasks concurrently.
4. Inspect each result and diff; do not trust summaries alone.
5. Resolve integration centrally.
6. Run the closest checks covering the combined boundaries.

Run a full suite only if combined work changed shared core behavior, crosses a
release boundary, or the user requested it.

More agents are not automatically faster. Stop dispatching when ownership
overlaps or briefing takes longer than the remaining work.

From tsergeytovarov/ai-settings, MIT.

Signals

Stars: 15
Forks: 4
Open issues: 2
Last commit (this folder): Jul 31, 2026 (2 months ago)
Last release: Oct 12, 2026 (16 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License: MIT
Repository: github.com/tsergeytovarov/ai-settings
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: GitHub github-graphql

More from tsergeytovarov

Cite this page

dispatching-parallel-agents skill. Agent Mart. https://agentmart.dev/skills/tsergeytovarov-dispatching-parallel-agents (last updated 2026-10-24)