Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add DevelopersGlobal/ai-agent-skills --skill task-decompositiongit clone --depth 1 https://github.com/DevelopersGlobal/ai-agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/developersglobal/ai-agent-skills/task-decomposition)<a href="https://agentmods.dev/skills/developersglobal/ai-agent-skills/task-decomposition"><img src="https://agentmods.dev/badge/skills/developersglobal/ai-agent-skills/task-decomposition/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/developersglobal/ai-agent-skills/task-decomposition"><img src="https://agentmods.dev/badge/skills/developersglobal/ai-agent-skills/task-decomposition.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00034 | $0.00659 |
| Opus 5.5 | $0.00014 | $0.00264 |
| Sonnet 5.5 | $0.00007 | $0.00132 |
| Haiku 4.5 | $0.00003 | $0.00066 |
Grade A, and why
task-decomposition scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Overview
Large tasks fail in large ways. Decomposed tasks fail in small, fixable ways. This skill breaks any feature or project into atomic tasks — each independently implementable, testable, and deployable.
When to Use
- Before starting any feature that takes more than half a day
- When a task feels overwhelming or unclear
- When multiple people need to work in parallel
Process
Step 1: Identify the Deliverable
- State what "done" looks like for the whole feature.
- List all the things that must be true when it's complete.
- Identify dependencies: what must exist before any task can start?
Verify: You can state the full feature goal in 2 sentences.
Step 2: Decompose Into Atomic Tasks
- Break the feature into tasks where each task:
- Can be completed in under 4 hours
- Has a single, clear output
- Can be verified independently
- Can be reverted without breaking other tasks
- Each task should be:
[verb] [noun] so that [outcome]- ✅ "Add rate limiting to /api/login so that brute-force is prevented"
- ❌ "Work on the login security stuff"
Verify: Every task is under 4 hours. Every task has a clear verify condition.
Step 3: Order and Parallelism
- Draw the dependency graph — what blocks what?
- Identify tasks that can be done in parallel.
- Sequence tasks so integration happens incrementally (not as one big bang at the end).
Verify: The dependency order is clear. No unnecessary sequential dependencies.
Step 4: Estimate and Adjust
- For each task, estimate: best case / worst case / expected.
- If any task's worst case > 1 day: decompose further.
- Total estimate sanity check: does it add up to a reasonable timeline?
Common Rationalizations (and Rebuttals)
| Excuse | Rebuttal |
|---|---|
| "I'll figure it out as I go" | Decomposition takes 30 minutes. Rework from poor planning takes days. |
| "It's too complex to break down" | Everything can be decomposed. Start with what you know, decompose the rest later. |
| "The tasks are too granular" | Granular tasks ship continuously. Coarse tasks ship never. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 75 lines · 34 tokens per session scan A 7ea286de79e4
task-decomposition is a skill published in the GitHub repository DevelopersGlobal/ai-agent-skills (215 stars, last pushed 5d ago), licensed MIT. It adds 34 tokens to every session and 659 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
to-tickets
Slice a spec or a ticket into tracer-bullet tickets with blocking edges, published through the repo's tracker: .specs/tickets/ .md by default.
to-spec
Turn the settled conversation into a spec (PRD), published through the repo's tracker: .specs/specs/ .md by default.
strapi
Expert Strapi headless CMS assistance covering content types, plugins, REST/GraphQL APIs, and custom controllers. Use when building self-hosted, customizable Node.js headless CMS backends.
setup
Configure this repo's issue tracker for the flow skills: GitHub, Jira, or local markdown (the default).
marker-client-report
Turn a project's completed Marker.io work into a client-ready report. Use when asked to write a weekly or monthly recap for a client, summarize what was shipped or fixed this period, or produce a status update for a website project from Marker.io. Read-only: it reads issues and writes a markdown report, it does not…
marker-triage
Triage a Marker.io website-feedback backlog at scale. Use when asked to go through, categorize, prioritize, deduplicate, or make sense of many Marker.io issues at once. Splits content fixes from functional bugs, groups by page, rates criticality, flags duplicates and repeat reporters, surfaces quick wins, lists…