Skip to content
Agentmart
Submit a tool

decomposing-to-issues

Use when breaking a spec or plan into tracker issues for execution — especially when different agents or sessions will pick the pieces up independently.

Skill by Roflochinsky

IndexedMIT
Write a review

Chalk Labs

Need help deploying decomposing-to-issues?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

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/roflochinsky-decomposing-to-issues

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.

We use your email only for this newsletter. Privacy

About

Use when breaking a spec or plan into tracker issues for execution — especially when different agents or sessions will pick the pieces up independently.

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. 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/decomposing-to-issues

What's inside

Skill name
decomposing-to-issues
Description
Use when breaking a spec or plan into tracker issues for execution — especially when different agents or sessions will pick the pieces up independently.
License
MIT
Instructions
26 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Decomposing to Issues

## Overview

Tasks larger than one context window get abandoned mid-way; tasks split by horizontal layer ("all backend", then "all frontend") serialize everything and integrate nothing until the end. Decompose into **vertical slices** sized for one fresh-context agent each, with contracts frozen up front so slices run in parallel.

## Rules

1. **Epic carries the frozen contracts.** Schema, API shapes, naming — pinned in the epic description before child issues are written. Children build against the contract, not against each other's unmerged code. Contract changes happen by editing the epic and notifying dependents, never silently inside a child.
2. **Vertical slices, tracer bullet first.** Each issue delivers something testable end-to-end or a self-contained layer of one feature path — not "all models for everything". The first issue is the thin universal blocker (e.g. migration + model), kept deliberately tiny so it merges fast and unblocks parallel tracks.
3. **Sized for one fresh agent.** An issue a new agent can complete from the issue text + spec/plan paths alone, without the conversation. If describing it needs "and then... and also..." — split it.
4. **Verification command inside each issue.** Every issue states how to prove it done (`pytest path/...`, `npm test -- x`). "Done" without a named check isn't a definition of done.
5. **Real dependencies only** (`bd dep add child blocker`), so `bd ready` reflects true execution order and everything else runs in parallel. Don't chain issues out of habit.
6. **A separate final verification issue**, owned by a non-implementer agent: walk the whole flow against the spec, file discrepancies as new issues rather than fixing inline.

## Anti-patterns

| Smell | Fix |
|---|---|
| Issues named "Backend", "Frontend", "Tests" | Horizontal layers. Re-cut by feature path. |
| Child issues referencing each other's future code | Missing frozen contract in the epic. |
| One issue with 8 checkboxes | 8 issues wearing a trenchcoat. Split. |
| "Verify everything works" as a checklist item of the last feature issue | Verification by the implementer. Separate issue, separate agent. |
| Every issue depends on the previous one | Fake serialization. Wire only real blockers. |

From Roflochinsky/workshopskills, MIT.

More from Roflochinsky/workshopskills

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jun 16, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Roflochinsky

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get decomposing-to-issues running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying decomposing-to-issues?

CallGet help
Cite this page

decomposing-to-issues skill. Agent Mart. https://agentmart.dev/skills/roflochinsky-decomposing-to-issues (last updated 2026-10-24)