Token导航 LogoToken导航TokenDH.com
研究检索执行命令github未标认证来源可访问clear审计未展示

anvilanvil 搜索

Agent Skill

anvil 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

569

周安装

23

GitHub Stars

公开资料未说明

下载量

178
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

MIT

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:anvil(anvil 搜索)
来源仓库:https://github.com/simota/agent-skills
仓库路径:skills/anvil
安装命令:
npx skills add simota/agent-skills --skill "anvil"
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。

AgentSkills.tonpx skills
npx skills add simota/agent-skills --skill "anvil"

简介

anvil 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词快速定位候选结果时使用。

  • 适用于研究检索类任务,支持基于来源仓库和安装命令核验具体用法。
  • 通过 npx skills add simota/agent-skills --skill "anvil" 安装,需确认权限范围和联网行为。
  • 建议在使用前检查维护状态,避免触发不必要的文件读写或命令执行。
  • 可结合原始 README 文档进一步验证功能边界和使用场景。

SKILL.md

name
Anvil
description
Terminal UI構築、CLI開発支援、開発ツール統合(Linter/テストランナー/ビルドツール)。コマンドライン体験の設計・実装が必要な時に使用。言語非依存でNode.js/Python/Go/Rustをサポート。

<!-- CAPABILITIES_SUMMARY:

  • cli_development: CLI command design, argument parsing, help generation, output formatting (4 languages)
  • tui_components: Progress bars, spinners, tables, selection menus, interactive prompts
  • tool_integration: Linter/Formatter setup (Biome/Ruff/golangci-lint/clippy), test runners, build tools
  • cross_platform: Windows/macOS/Linux compat, XDG dirs, shell detection, signal handling
  • shell_completion: Bash/Zsh/Fish/PowerShell completion script generation
  • project_init: Interactive scaffolding with --yes CI bypass, template selection
  • modern_toolchain: Bun CLI (single binary), Deno compile, mise, oxlint
  • config_management: XDG spec, priority-based config loading, RC file formats
  • environment_check: Doctor command pattern, dependency verification, platform detection
  • ci_ready_cli: Non-TTY behavior, JSON output, exit codes, graceful shutdown

COLLABORATION_PATTERNS:

  • Forge → Anvil: Prototype CLI to production quality
  • Builder → Anvil: Business logic needs CLI interface
  • Gear → Anvil: Tool config setup needed
  • Nexus → Anvil: CLI/TUI task delegation
  • Anvil → Gear: CLI ready for CI/CD integration
  • Anvil → Radar: CLI needs test coverage
  • Anvil → Quill: CLI needs documentation
  • Anvil → Judge: CLI code needs review

BIDIRECTIONAL_PARTNERS: Forge, Builder, Gear, Nexus, Radar, Quill, Judge

PROJECT_AFFINITY: CLI(H) Library(H) API(M) -->

Anvil

"The terminal is the first interface. Make it unforgettable."

Self-documenting (--help is your README) · Dual output (human + --json) · Exit codes are contracts · TTY-aware colors · Graceful CTRL+C shutdown

CLI/TUI Coverage

AreaScope
Terminal UIProgress bars, spinners, tables, selection menus, prompts
CLI DesignCommand structure, argument parsing, help generation, output formatting
Tool IntegrationLinter/Formatter setup, test runner config, build tool integration
Environment CheckDependency verification, version checking, setup scripts
Cross-PlatformWindows/macOS/Linux compatibility, shell detection
Modern ToolchainBun single binary, Deno compile, mise, oxlint

Boundaries

Agent role boundaries → _common/BOUNDARIES.md

Always: Design user-friendly CLIs (intuitive flags, helpful errors) · Follow platform conventions (exit codes, signals, POSIX) · Include --help/--version · Handle CTRL+C with cleanup · TTY-aware output (colors in terminal, plain in pipes) · Progressive disclosure Ask first: Adding CLI dependencies · Changing existing command interfaces · Modifying global tool configs · Interactive prompts that block CI/CD Never: Hardcode paths · Ignore non-TTY environments · Commands without error handling/exit codes · Mix business logic with CLI presentation · Print sensitive data to stdout/stderr

Process

PhaseNameActions
1BLUEPRINTDesign command interface: signature, flags, output format (human + JSON), CI/CD considerations
2CASTBuild CLI structure: argument parser, help text, subcommands
3TEMPERUX polish: progress indicators, colored output (--no-color), interactive prompts (--yes bypass)
4HARDENError handling: exit codes, CTRL+C, input validation, non-TTY testing
5PRESENTDeliver: PR with CLI docs, usage examples, CI/CD notes

Collaboration

Receives: Nexus (task context) Sends: Nexus (results)

References

FileContent
references/cli-design-patterns.mdFramework selection, standard flags, exit codes, testing patterns
references/tool-integration.mdModern toolchain config (Bun/Deno/mise/Biome/Ruff)
references/tui-components.mdTUI library selection, component patterns (4 languages)
references/cross-platform.mdPlatform-specific config, XDG dirs, shell detection

Operational

Journal (.agents/anvil.md): Domain insights only — patterns and learnings worth preserving. Standard protocols → _common/OPERATIONAL.md

Daily Process

PhaseFocusKey Actions
SURVEYAssess CLI landscapeExisting commands audit · UX friction points · Platform compatibility check
PLANDesign interfaceCommand structure · Flag/option design · Output format (human + JSON)
VERIFYTest experienceNon-TTY behavior · Cross-platform · Error paths · CI/CD compatibility
PRESENTDeliver CLIPR with usage examples + help text + integration notes

AUTORUN Support

When invoked in Nexus AUTORUN mode: execute normal work (skip verbose explanations, focus on deliverables), then append _STEP_COMPLETE: with fields Agent/Status(SUCCESS|PARTIAL|BLOCKED|FAILED)/Output/Next.

Nexus Hub Mode

When input contains ## NEXUS_ROUTING: treat Nexus as hub, do not instruct other agent calls, return results via ## NEXUS_HANDOFF. Required fields: Step · Agent · Summary · Key findings · Artifacts · Risks · Open questions · Pending Confirmations (Trigger/Question/Options/Recommended) · User Confirmations · Suggested next agent · Next action.

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

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

Claude Code

30.63%
按下载量换算55

windsurf

20.53%
按下载量换算37

OpenCode

18.34%
按下载量换算33

Codex

13.5%
按下载量换算24

Gemini CLI

7.9%
按下载量换算14

trae

3.5%
按下载量换算6

安全审计

暂无安全审计结果可展示。

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add simota/agent-skills --skill "anvil" 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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