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 agentmods add skills/dixus/claudeframework/decomposenpx skills add dixus/claudeframework --skill decomposegit clone --depth 1 https://github.com/dixus/claudeframeworkWhat 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 | $0.00036 | $0.01795 |
| Opus 5 | $0.00018 | $0.00898 |
| Sonnet 5 | $0.00007 | $0.00359 |
| Haiku 4.5 | $0.00004 | $0.00179 |
Grade A, and why
decompose 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 2d 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 — 222 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Decompose concept documents in .claude/input/ into independent, shippable PRDs.
$ARGUMENTS is an optional name or focus hint. The primary source is .claude/input/.
Step 1 — Ingest all input materials
Read every file in .claude/input/ — documents, images, screenshots, PDFs, sketches, markdown, plain text. Treat them collectively as a single concept brief.
Also read:
CLAUDE.mdfor project context, tech stack, and conventions- All files in
.claude/context/for long-lived project references (schemas, API docs, glossaries)
Build a mental model of what the concept describes: the overall goal, the user-facing features, the technical components, and any constraints or priorities mentioned.
Step 2 — Identify feature boundaries
Analyze the concept and identify independent features — units of work that:
- Have a clear, single purpose — each delivers one user-visible capability or one infrastructure change
- Can be implemented without the others existing — no hard runtime dependency on another feature from this decomposition (soft dependencies like "uses the same data model" are fine)
- Can be reviewed and tested in isolation — each has its own validation criteria
- Touch mostly different files — overlap of 1-2 shared files (e.g., a route config, a nav menu) is acceptable, but features that modify the same core logic in conflicting ways should be merged or sequenced
For each candidate feature, note:
- A short name (kebab-case, suitable for filenames)
- Which parts of the concept it covers
- Which specific files it will likely create or modify (list paths —
/fleetuses these for conflict analysis) - Any dependencies on other features from this decomposition
Step 3 — Dependency analysis and ordering
Build a dependency graph between the identified features:
- Independent — can run in any order, no dependency
- Soft dependency — benefits from another feature existing first (e.g., a dashboard that shows data from a new API), but can be implemented with stubs or existing data
- Hard dependency — cannot be implemented until another feature ships (e.g., a feature that extends a new data model created by another feature)
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.
- 2d ago First seen · 222 lines · 36 tokens per session scan A c15428fadeee
decompose is a skill published in the GitHub repository dixus/claudeframework (10 stars, last pushed 4mo ago), licensed MIT. It adds 36 tokens to every session and 1,795 once invoked, about $0.0002 per session on Opus 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.