Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计通过

pasta-decompose面食分解

Agent Skill

pasta-decompose 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

275

周安装

11

GitHub Stars

9

下载量

89
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:pasta-decompose(面食分解)
来源仓库:https://github.com/florianbuetow/claude-code
仓库路径:skills/pasta-decompose
安装命令:
npx skills add https://github.com/florianbuetow/claude-code --skill pasta-decompose
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/florianbuetow/claude-code --skill pasta-decompose

简介

用于查找、检索和筛选相关信息。pasta-decompose 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

  • 适合根据关键词或任务场景快速定位候选结果。
  • 可结合来源仓库和原始 README 核验具体用法。
  • 安装前建议确认权限范围和维护状态。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。
  • 注意是否会触发联网或文件读写操作。

SKILL.md

PASTA Stage 3: Application Decomposition

Break the application into its constituent components and analyze trust boundaries, user roles, privilege levels, and data sensitivity at each layer. This produces the component map that Stages 4-7 use to identify where threats and attacks focus.

Supported Flags

Read ../../shared/schemas/flags.md for the full flag specification. Key behaviors:

FlagStage 3 Behavior
--scopeDefault changed. Scans middleware, auth modules, role definitions, permission configs, database schemas.
--depth quickComponent inventory from project structure and imports only.
--depth standardFull decomposition + role-permission matrix + data classification.
--depth deepStandard + cross-component data flows, inter-service auth, shared resource access patterns.
--depth expertDeep + formal trust boundary analysis with privilege escalation path mapping.
--severityNot applicable at this stage.

Framework Context

Read ../../shared/frameworks/pasta.md, Stage 3 section. PASTA is SEQUENTIAL. Stage 3 consumes Stage 2 output and feeds Stage 4.

Prerequisites

Required: Stage 2 output -- entry point inventory, DFD, technology stack, and external dependencies. If unavailable, warn and proceed with assumptions.

Workflow

Step 1: Determine Scope

Parse --scope flag (default: changed). Prioritize: middleware, auth modules, role/permission definitions, policy files (RBAC, ABAC, ACL), database schemas, service definitions, module boundaries.

Step 2: Identify Components

Decompose into functional units: authentication, authorization, core business logic, data access layer, API layer, background processing, admin/management, integration layer (third-party clients, webhooks), and infrastructure (logging, caching, sessions).

Step 3: Map Roles and Permissions

For each component: identify user roles (anonymous, authenticated, admin, service account), permission model (RBAC/ABAC/ACL), enforcement location (middleware, decorators, inline), default permissions, and permission gaps.

Step 4: Map Trust Boundaries

Identify where trust levels change: external-to-app, unauthenticated-to-authenticated, user-to-admin, service-to-service (mTLS, API keys, JWT), app-to-datastore, and app-to-third-party.

Step 5: Classify Data

For each data entity: Public (marketing, public APIs), Internal (analytics, internal docs), Confidential (PII, credentials, tokens), Restricted (payment data, health records, encryption keys).

Step 6: Document Auth/Authz Flows

Trace end-to-end: credential validation, session/token issuance, per-request permission checks, and bypass paths (debug endpoints, feature flags, test accounts).

Analysis Checklist

  1. What trust boundaries does data cross between components?
  2. Which components run with elevated privileges?
  3. How are service-to-service calls authenticated?
  4. Where does data sensitivity change (encryption, masking, aggregation)?
  5. Are there shared databases or caches across trust boundaries?
  6. What is the default permission level for new users or services?
  7. Are there components that lack authorization checks entirely?
  8. Do admin interfaces share authentication with user-facing interfaces?

Output Format

Stage 3 produces a Component Inventory with Trust Boundaries. ID prefix: PASTA (e.g., PASTA-S3-001).

## PASTA Stage 3: Application Decomposition

### Component Inventory
| ID | Component | Function | Trust Level | Auth Mechanism |
|----|-----------|----------|-------------|---------------|
| C-01 | Auth Service | Login, registration, JWT | High | N/A (is the auth) |
| C-02 | API Gateway | Request routing | Medium | JWT validation |
| C-03 | Payment Module | Charge processing | High | JWT + role check |

### Role-Permission Matrix
| Role | C-01 Auth | C-02 API | C-03 Payment | C-04 Admin |
|------|----------|---------|-------------|-----------|
| Anonymous | register, login | public endpoints | none | none |
| User | profile, logout | CRUD own data | pay, history | none |
| Admin | manage users | all endpoints | refunds | full access |

### Data Classification
| Data Entity | Classification | Components | Encrypted Rest/Transit |
|------------|---------------|------------|----------------------|
| Credentials | Restricted | Auth Service | Hashed / TLS |
| Payment tokens | Restricted | Payment | Yes / TLS |

### Trust Boundaries
| Boundary | From | To | Mechanism | Validated |
|----------|------|----|-----------|-----------|
| TB-01 | Internet | API Gateway | HTTPS | TLS |
| TB-02 | Gateway | Auth Service | Internal HTTP | JWT |

### Authorization Gaps
[Components or endpoints without proper authorization enforcement]

Findings follow ../../shared/schemas/findings.md with:

  • metadata.tool: "pasta-decompose", metadata.framework: "pasta", metadata.category: "Stage-3"

Next Stage

Stage 4: Threat Analysis (pasta-threats). Pass the Component Inventory, Role-Permission Matrix, Data Classification, and Trust Boundaries. Stage 4 cross-references with real-world threat intelligence to identify relevant threats.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.63%
按下载量换算34

Claude

31.59%
按下载量换算28

Cursor

18.41%
按下载量换算16

Gemini CLI

9.33%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills