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 brody-0125/my-claude-skills --skill numericalgit clone --depth 1 https://github.com/brody-0125/my-claude-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/brody-0125/my-claude-skills/numerical)<a href="https://agentmods.dev/skills/brody-0125/my-claude-skills/numerical"><img src="https://agentmods.dev/badge/skills/brody-0125/my-claude-skills/numerical/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/brody-0125/my-claude-skills/numerical"><img src="https://agentmods.dev/badge/skills/brody-0125/my-claude-skills/numerical.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 | $0.00107 | $0.04134 |
| Opus 5.5 | $0.00043 | $0.01654 |
| Sonnet 5.5 | $0.00021 | $0.00827 |
| Haiku 4.5 | $0.00011 | $0.00413 |
Grade A, and why
numerical 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 2mo 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 — 355 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Numerical — Numerical Computing Verification & Optimization Agent
An agent that verifies correctness and optimizes performance of numerical computing code through Analyze → Verify → Optimize workflow
Role
A workflow agent specializing in numerical computing for Python and Dart projects. It automatically detects the project language (Python/Dart/Mixed), numeric libraries (NumPy, SciPy, CuPy, dart_tensor, etc.), and provides expert-level verification and optimization of array/tensor operations, floating-point arithmetic, broadcasting, memory layout, SIMD utilization, and GPU processing.
Core Principles
- Lazy-load context documents per Phase to ensure verification thoroughness
- Sequential execution of Analyze → Verify → Optimize
- Repeat Verify loop the number of times specified by the user (Ralph-style)
- Automatically adjust verification level based on computational complexity (Tiered Verification)
- IEEE 754 compliance as the baseline for all floating-point analysis
Quick Start (Zero-Config)
Phase 0 자동으로 모든 설정을 완료하므로 사용자 개입이 필요 없다:
1. Project Discovery — pyproject.toml / pubspec.yaml 분석 → 언어, 라이브러리, GPU 지원 자동 감지
2. Numeric Profile — dtype 사용 패턴, 배열 차원, 연산 유형 자동 프로파일링
3. Tool Detection — pytest/ruff/mypy/dart analyze 등 검증 도구 자동 감지
4. Hooks Installation — lint-on-edit, secret-guard, test-quality-gate 자동 설치
첫 실행 시 추가 프롬프트 없이 위 4단계가 순차적으로 실행된다. 감지된 설정을 변경하려면 해당 파일을 직접 편집하면 된다:
- 정적 분석 도구:
.numerical/analysis-tools.txt(줄 단위, 삭제 시 재감지) - Hooks:
.claude/settings.json의hooks섹션 (삭제 시 재설치)
Phase Workflow Diagram
┌──────────────────────────────────────────────────────────────────────┐
│ numerical │
└──────────────────────────────────────────────────────────────────────┘
│
▼
┌──────────────────────────┐
│ Phase 0: Discovery │
│ • 언어/라이브러리/GPU 감지 │
│ • 수치 프로파일 캐시 저장 │
└────────────┬─────────────┘
│
┌────────────▼─────────────┐
│ Phase 1: Analyze │
│ • 수치 연산 패턴 분석 │
│ • dtype/shape 추적 │
│ • 정밀도 위험 감지 │
└────────────┬─────────────┘
│
┌──── dry-run? ────┴──────────────────┐
│ │
┌────▼────┐ ┌───────────▼───────────┐
│ HALT │ │ Phase 2: Verify │
│ (리포트만)│ │ • 부동소수점 정합성 검증 │
└─────────┘ │ • 브로드캐스팅 규칙 검증 │
│ • 테스트 케이스 검증 │
│ • 에지 케이스 분석 │
└───────────┬───────────┘
│
┌───────────────────────────────▼───────────────────────────────┐
│ Phase 3: Optimize │
│ ┌─────────────────────────────────────────────────────────┐ │
│ │ • SIMD 정렬/벡터화 최적화 제안 │ │
│ │ • GPU 메모리 관리 최적화 │ │
│ │ • 메모리 레이아웃 (C/F-contiguous) 최적화 │ │
│ │ • 알고리즘 수치 안정성 개선 │ │
│ └─────────────────────────┬───────────────────────────────┘ │
│ │ │
│ ┌────────────────▼────────────────┐ │
│ │ 종료 조건 확인 │ │
│ │ • 위반 0개? │ │
│ │ • 정밀도 목표 달성? │ │
│ │ • 동일 이슈 3회 반복? │ │
│ │ • max loop 도달? │ │
│ └────────────────┬────────────────┘ │
│ │ │
│ ┌─── EXIT ─────────┴─────── CONTINUE ───┐ │
│ │ │ │
│ │ Loop N++ (재검증) │
│ │ │ │
└─────────┼────────────────────────────────────────┘ │
│ │
▼ │
┌──────────────────────────┐ │
│ Complete │◄────────────────────────────────────┘
│ • 분석 리포트 출력 │
│ • PROGRESS.md 기록 │
└──────────────────────────┘
What ships with it
20 files 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.
- references/broadcasting-rules.md 5.1 KB
- references/dart-tensor-conventions.md 7.4 KB
- references/floating-point-guide.md 6.0 KB
- references/gpu-memory-guide.md 7.3 KB
- references/numpy-conventions.md 6.1 KB
- references/simd-alignment-guide.md 5.4 KB
- resources/analyze-protocol.md 5.8 KB
- resources/error-playbook.md 10 KB
- resources/optimize-protocol.md 6.2 KB
- resources/project-discovery-protocol.md 4.8 KB
- resources/verification-tiers.md 5.2 KB
- resources/verify-protocol.md 7.7 KB
- scripts/_common.sh 4.5 KB runs code
- scripts/discover-project.sh 5.7 KB runs code
- scripts/setup-hooks.sh 1.6 KB runs code
- scripts/verify-numeric.sh 9.3 KB runs code
- templates/analysis-report-template.md 1.6 KB
- templates/hooks-config.json 2.5 KB
- templates/optimization-report-template.md 1.4 KB
- templates/progress-template.md 1.2 KB
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.
- 2mo ago First seen · 355 lines · 107 tokens per session scan A 453b34c4e227
numerical is a skill published in the GitHub repository brody-0125/my-claude-skills (2 stars, last pushed 8mo ago), licensed MIT. It adds 107 tokens to every session and 4,134 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
milp-modeling-gurobi
When the user wants to build, solve, and debug mixed-integer linear programs in Python with Gurobi — creating variables, writing constraint-builder functions, setting objectives and parameters, handling solver status, and extracting solutions safely. Also use when the user mentions "gurobipy," "build a MIP model,"…
numpy-vectorization-for-optimization
When the user wants to remove slow Python loops from metaheuristic or optimization code using NumPy — population-level operations, batch fitness evaluation, distance matrices, broadcasting, argsort/argpartition idioms, defaultrng, and memory layout. Also use when the user mentions "vectorize," "numpy broadcasting,"…
jupyter-notebook
Iterative Python via live Jupyter kernel (hamelnb).
matlab
Builds, reviews, migrates, and plans MATLAB or GNU Octave numerical workflows. Use for arrays, tabular/time data, tests, projects, graphics, MAT files, and explicit Python interoperability.
pennylane
Builds and differentiates PennyLane quantum circuits, hybrid PyTorch or JAX models, molecular VQE and QAOA workflows. Use for variational quantum algorithms, quantum machine learning, simulator validation, and moving validated circuits to provider plugins. For hardware-specific compilation use qiskit or cirq; for…
cuopt-numerical-optimization-api
LP, MILP, and QP (beta) with cuOpt — Python, C, and CLI. Use when the user is solving LP, MILP, or QP with any cuOpt interface.