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/d-oit/rust-2026-template/task-decompositionnpx skills add d-oit/rust-2026-template --skill task-decompositiongit clone --depth 1 https://github.com/d-oit/rust-2026-templateWhat 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 | $0.00056 | $0.00713 |
| Opus 5 | $0.00028 | $0.00357 |
| Sonnet 5 | $0.00011 | $0.00143 |
| Haiku 4.5 | $0.00006 | $0.00071 |
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 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task Decomposition
Decompose high-level objectives into manageable, testable sub-tasks.
When to Use
- Complex requests with multiple components
- Multi-phase projects requiring coordination
- Tasks benefiting from parallel execution
Framework
1. Requirements Analysis
Extract: Primary objective, implicit requirements, constraints, success criteria.
2. Goal Hierarchy
Main Goal
├─ Sub-goal 1 → Tasks 1.1, 1.2
├─ Sub-goal 2 → Task 2.1
└─ Sub-goal 3
Atomic Criteria: Single action, defined inputs/outputs, one agent, testable.
3. Dependency Mapping
- Sequential: A → B → C
- Parallel: A, B, C (independent)
- Converging: A, B, C → D
4. Success Criteria
Define: Inputs, outputs, quality standards.
Process
Step 1: Understand Goal
Request: [Original]
Goal: [Main objective]
Type: [Implementation/Debug/Refactor]
Complexity: [Simple/Medium/Complex]
Step 2: Identify Components (3-7)
Step 3: Decompose Components
Step 4: Map Dependencies
Step 5: Assign Priorities
- P0 (Critical), P1 (Important), P2 (Nice-to-have)
Step 6: Estimate Complexity
- Low (<30min), Medium (30min-2hr), High (>2hr)
Patterns
Layer-Based (Rust)
- Data types / schemas
- Business logic
- API / CLI interface
- Integration tests
- Documentation
Feature-Based
- Core (MVP)
- Error handling (
thiserror/anyhow) - Performance (benches)
- Integration
- Testing
- Docs
Problem-Solution
- Reproduce
- Diagnose
- Design
- Fix
- Verify
- Prevent
Quality Checklist
- Atomic and actionable tasks
- Dependencies identified
- Measurable success criteria
- Appropriate complexity estimates
- No task >4 hours
- Parallelization opportunities found
Rationalizations
| Rationalization | Reality |
|---|---|
| "I already know what needs to be done, no need to decompose" | Skipping decomposition leads to scope creep, missed dependencies, and rework. |
| "The task is too simple to break down" | Even simple tasks benefit from explicit success criteria and dependency mapping. |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 127 lines · 56 tokens per session scan A b83e5f409afc
task-decomposition is a skill published in the GitHub repository d-oit/rust-2026-template (10 stars, last pushed 3d ago), licensed MIT. It adds 56 tokens to every session and 713 once invoked, about $0.0003 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
harbor
Run Terminal-Bench or any Harbor dataset with Ante as the agent, using the Harbor adapter in the ante repo's ante-harbor/ directory. Use when the user wants to benchmark Ante on Harbor or reproduce Ante's published eval results.
check-model
Check model compatibility with xinfer before loading. Validates config.json, weight tensor shapes and naming, quantization format correctness, and multi-rank (tensor-parallel) divisibility. Use when the user asks to check, validate, audit, or verify a model will load correctly — from a HuggingFace URL/config, local…
test-model
Test LLM models served by xinfer for correctness, output quality, and performance. Use when the user asks to test, benchmark, validate, or verify models — either from a local folder path or HuggingFace model IDs. Supports all xinfer-compatible formats: BF16, FP8, MXFP4, NVFP4, GGUF, GPTQ, AWQ, ISQ, Dense, MoE, and…
sivtr-memory
Retrieve shared local work memory: terminal activity, AI conversation history, prior decisions, validation evidence, debugging trails, recaps, and handoff context. Use before asking the user to repeat local work context.
skills
Verify sivtr environment and diagnose issues. Run when sivtr commands fail or return empty results.
github-actions-pat
用 Personal Access Token 通过 GitHub REST API 分析 Actions CI 的失败 run/job/step、拉取日志、轮询运行状态、做 PR code review，并驱动 fix → push → watch → iterate 的闭环。Token 只从 GITHUBTOKEN 环境变量读取，不落盘、不回显。适合在 bifrost remote / CI 调度器 / 无头环境里跑。仓库锁定在 AGENTS.md 或调用处通过 GHREPO 显式指定。.