AIGCLIST
基准评分
SKILL.md 在 Agent 就绪度与 AI 可见性上的得分 AI 就绪度和 GEO Score 是 VibeLaunch 在提交后生成的平台评估。
决策摘要
AI developers and agent builders working within the Claude Code ecosystem
Discovering, evaluating, and installing community-contributed AI agent skills from GitHub repositories
适合
- Claude Code developers seeking community-contributed skills
- AI agent builders exploring API exploration and self-improvement tools
- Bilingual EN/ZH development teams needing consistent tool evaluation
注意
- Community-driven content lacks independent quality verification
- Nascent platform with limited catalog depth and early-stage community
- Skills depend on upstream GitHub repositories remaining actively maintained
概述
评价 (0)
还没有评价。成为第一个评价的人！
评分构成
编辑评分由哪些维度构成，每项附判断依据。 AI 就绪度和 GEO Score 是 VibeLaunch 在提交后生成的平台评估。
Information quality
Skill pages provide structured information with AI summaries and full README content, but there is no independent editorial review or verification of vendor claims. Information quality depends entirely on upstream repository maintainers.
Skill listings include AI summaries, README from GitHub, and installation instructions (c003). Vendor claims about capabilities like self-improvement (c005) and progressive API exploration (c004) are presented without independent validation.
Ease of use
Clean, consistent skill page structure with multiple installation methods and bilingual support. Navigation is straightforward with category browsing. No registration or authentication required to access the catalog.
Copy-ready git clone commands for HTTPS, SSH, and CLI methods on each page (c007). Consistent layout across skill listings (c003). Bilingual paths readily accessible (c002).
Feature depth
The platform functions primarily as a structured catalog with browsing and installation. It lacks advanced discovery features such as faceted search, skill comparison, ratings, or automated compatibility checking.
Core features center on listing, summarization, and GitHub installation (c003, c007). No evidence of search filtering, quality scoring, automated testing, or integration validation beyond basic skill page structure.
Workflow fit
Directly targets Claude Code developers with skills designed for terminal-based agent workflows. GitHub-native installation aligns with developer tooling conventions. Limited to a single agent ecosystem.
All cataloged skills target Claude Code (c001, c008). Installation follows standard developer workflows via git clone (c007). Seamless integration emphasized as a design principle (c009). Narrow ecosystem scope limits broader applicability.
Reliability
Nascent platform with early-stage community metrics. Skills are community-contributed without quality gates. Upstream GitHub dependency creates a single point of failure if repositories become unmaintained.
Openapi-mcp-proxy shows 12 stars and 5 forks as of early 2026 (c010). No evidence of automated testing, skill validation pipelines, or maintainer accountability mechanisms. Skills depend entirely on upstream repo availability (c007).
Value
Free access to a growing catalog of AI agent skills with no paywall or registration barrier. The primary cost is developer time spent evaluating unverified skills, which is reasonable for the target audience of open-source-comfortable developers.
Platform is freely browsable with public skill pages (c001, c003). Skills sourced from public GitHub repositories with open-source licensing (c007). No evidence of paid tiers, subscription requirements, or usage limits.
评分反映可查证的产品资料，不代表实际使用效果保证。
Agent 就绪度
评估 Agent 能否通过产品的官方信息理解产品，并重建一条有文档依据的工作流程。
Automated agent-readiness assessment of https://agent-skills.cc: 8 of 22 checks verified across 5 fetched pages. No substantial machine interface is documented — agents can understand and cite the product but not operate it. Absent: quickstart, api_reference, request_examples, response_examples, rate_limits, version_information.
就绪度维度
| 评估维度 | 得分 |
|---|---|
| 文档质量 | 38 |
| 执行结果可验证性 | 10 |
| 机器接口 | 25 |
| 项目定位清晰度 | 75 |
| 资源可发现性 | 100 |
| 工作流完整度 | 69 |
对 Agent 有帮助的部分
- docs: verified during this run
- llms txt: verified during this run
- sitemap: verified during this run
- agent tooling artifacts: verified during this run
- authentication: verified during this run
- cli: verified during this run
Agent 受阻的部分
- No quickstart signal matched across 5 fetched pages.
- No api reference signal matched across 5 fetched pages.
- No version information signal matched across 5 fetched pages.
- No changelog signal matched across 5 fetched pages.
- No cli non interactive signal matched across 5 fetched pages (a CLI is documented, but not this property).
- No cli structured output signal matched across 5 fetched pages (a CLI is documented, but not this property).
| 检查项 | 状态 | 详情 |
|---|---|---|
| 理解产品1/5 已核验 | ||
| 产品文档 | 已核验 | Developer/documentation pages reachable from the entry page (e.g. https://agent-skills.cc/claude-skills/documentation). |
| 快速开始 | 未在本次官方来源链中找到 | |
| API 参考 | 未在本次官方来源链中找到 | |
| 请求示例 | 不适用于该产品 | No API surface is offered; the product is operated through agent tooling instead. |
| 响应示例 | 不适用于该产品 | No API surface is offered; the product is operated through agent tooling instead. |
| 连接接口2/4 已核验 | ||
| SDK | 未在本次官方来源链中找到 | |
| MCP 接口 | 已核验 | Probe matched on https://agent-skills.cc/skills/openclaw-openclaw: /model context protocol|\bmcp\b(?!-)/. |
| Webhooks | 未在本次官方来源链中找到 | |
| 认证文档 | 已核验 | Probe matched on https://agent-skills.cc/skills/openclaw-openclaw: /api key|bearer|oauth|access token|authen/. |
| 执行工作流2/6 已核验 | ||
| 命令行工具 | 已核验 | Probe matched on https://agent-skills.cc/claude-skills/documentation: /\bcli\b|command[- ]line interface|npm (i/. |
| 非交互式命令 | 未在本次官方来源链中找到 | |
| 命令行结构化输出 | 未在本次官方来源链中找到 | |
| 结构化导入与导出 | 未在本次官方来源链中找到 | |
| 成功状态验证 | 未在本次官方来源链中找到 | |
| 智能体工具产物 | 已核验 | Multiple agent tooling artifacts documented: agent instruction files (AGENTS.md / CLAUDE.md / skills) referenced on https://agent-skills.cc/; code distribution via git clone documented on https://agent-skills.cc/skills/openclaw-openclaw; named slash-command skills (≥2 distinct) documented on https://agent-skills.cc/. |
| 维护与排错0/4 已核验 | ||
| 错误文档 | 部分可用 | Weak signal on the entry page only: /error (codes?|handling|responses?)|4xx|5/. |
| 速率限制 | 不适用于该产品 | No API surface is offered; the product is operated through agent tooling instead. |
| 版本信息 | 未在本次官方来源链中找到 | |
| 更新日志 | 未在本次官方来源链中找到 | |
| 发现与验证3/3 已核验 | ||
| llms.txt | 已核验 | llms.txt published at the site root (87 lines). |
| 站点地图 | 已核验 | sitemap.xml reachable and lists site pages. |
| 智能体原生定位 | 已核验 | Agent-native positioning with a concrete operational path: "The site explicitly references the SKILL.md open standard and provides installation commands (git clone) for agent skills, offering a concrete operational path for AI agents.". |
官方证据
审计信息
- 评测时间
- 2026年8月30日
- 评测基准
- agent-readiness-v1
- 读取页面
- 5
- 来源深度
- 1
本审计从一个入口 URL 及其经过验证的官方来源链评估文档所支持的可操作性。AIGCLIST 未注册、登录、购买、执行或测试该产品的运行可靠性。
证据核查
关于该工具的公开声明，每条均标注核验状态与引用来源。
skills/nyudenkov-openapi-mcp-proxy已验证4agent-skills.cc已验证核验于 2026年7月14日
Agent Skills (agent-skills.cc) is a web-based community registry that catalogs AI agent skills, primarily targeting the Claude Code ecosystem.
The openapi-mcp-proxy skill uses progressive exploration of large OpenAPI specifications, avoiding full-context dumping into the AI's working memory by letting developers navigate endpoints and data models incrementally.
Claude Code is described as an agentic coding tool that operates in the terminal, understands codebases, and executes routine development tasks through natural language commands.
The platform characterizes openapi-mcp-proxy as GPS navigation for API documentation that eliminates needle-in-a-haystack manual searches.
https://agent-skills.cc/skills/nyudenkov-openapi-mcp-proxyzh/skills/garyblankenship-SKILL.md已验证4agent-skills.cc已验证核验于 2026年7月14日
Agent Skills (agent-skills.cc) is a web-based community registry that catalogs AI agent skills, primarily targeting the Claude Code ecosystem.
Each skill listing includes an AI-generated summary, GitHub installation instructions, full README content, and links to related skills for structured discovery.
The SKILL.md skill enables Claude AI assistants to self-improve by automatically learning from documentation, codebases, or skill marketplaces without requiring manual updates.
The platform emphasizes seamless integration of listed skills into the Claude Code agent environment as a core design principle.
https://agent-skills.cc/zh/skills/garyblankenship-SKILL.mdCurated Skills Marketplace | 1,000+ Claude Code Skills | Agent Skills已验证3agent-skills.cc已验证核验于 2026年8月30日
The entry page was fetched and analyzed for machine-interface signals (title, headings, developer links, keyword probes).
Agent tooling artifacts observed: agent instruction files (AGENTS.md / CLAUDE.md / skills) referenced on https://agent-skills.cc/; code distribution via git clone documented on https://agent-skills.cc/skills/openclaw-openclaw; named slash-command skills (≥2 distinct) documented on https://agent-skills.cc/.
Agent-native positioning with a concrete operational path: "The site explicitly references the SKILL.md open standard and provides installation commands (git clone) for agent skills, offering a concrete operational path for AI agents.".
https://agent-skills.cc/zh/skills/nyudenkov-openapi-mcp-proxy已验证3agent-skills.cc已验证核验于 2026年7月14日
The platform provides bilingual content in English and Simplified Chinese through dedicated zh/ locale paths with culturally adapted skill descriptions.
Skills can be installed from GitHub repositories using HTTPS, SSH, or GitHub CLI methods, with copy-ready commands provided on each skill page.
The openapi-mcp-proxy skill repository, created on 2026-01-29, has community engagement of 12 stars and 5 forks on GitHub.
https://agent-skills.cc/zh/skills/nyudenkov-openapi-mcp-proxyzh/skills/garyblankenship-SKILL.md已验证3agent-skills.cc已验证核验于 2026年7月14日
The platform provides bilingual content in English and Simplified Chinese through dedicated zh/ locale paths with culturally adapted skill descriptions.
Each skill listing includes an AI-generated summary, GitHub installation instructions, full README content, and links to related skills for structured discovery.
The platform emphasizes seamless integration of listed skills into the Claude Code agent environment as a core design principle.
https://agent-skills.cc//zh/skills/garyblankenship-SKILL.mdskills/newbdez33-claude-code-skills-reference厂商声明2agent-skills.cc厂商声明核验于 2026年7月14日
The claude-code-skills-reference skill provides instant answers for programming language syntax questions and automatically remembers frequently used code templates.
Users interact with skills through natural language descriptions to receive coding answers, syntax references, and API navigation assistance.
https://agent-skills.cc/skills/newbdez33-claude-code-skills-referencehttps://agent-skills.cc/llms.txt已验证1agent-skills.cc已验证核验于 2026年8月30日
llms.txt is published at the site root and readable.
https://agent-skills.cc/llms.txthttps://agent-skills.cc/sitemap.xml已验证1agent-skills.cc已验证核验于 2026年8月30日
sitemap.xml is reachable and lists site pages.
https://agent-skills.cc/sitemap.xmlClaude Code Skills Documentation - API Docs, README, OpenAPI | Agent Skills已验证1agent-skills.cc已验证核验于 2026年8月30日
A documentation surface is reachable at https://agent-skills.cc/claude-skills/documentation.
https://agent-skills.cc/claude-skills/documentationHot Claude Code Skills on GitHub (Sorted by Stars/Forks) | Agent Skills已验证1agent-skills.cc已验证核验于 2026年8月30日
A quick-start / agent-skills documentation page is reachable at https://agent-skills.cc/claude-skills/hot.
https://agent-skills.cc/claude-skills/hotopenclaw-development-Claude Code Skill | Agent Skills已验证1agent-skills.cc已验证核验于 2026年8月30日
A quick-start / agent-skills documentation page is reachable at https://agent-skills.cc/skills/openclaw-openclaw.
https://agent-skills.cc/skills/openclaw-openclawawesome-chatgpt-prompts-data-ai-Claude Code Skill | Agent Skills已验证1agent-skills.cc已验证核验于 2026年8月30日
A quick-start / agent-skills documentation page is reachable at https://agent-skills.cc/skills/f-awesome-chatgpt-prompts.
https://agent-skills.cc/skills/f-awesome-chatgpt-prompts决策核对台
在依赖该产品或访问官网前，最值得先确认的问题。
Agent Skills is a community-driven web registry at agent-skills.cc that catalogs AI agent skills compatible with Claude Code. It provides structured listings with AI-generated summaries, full README documentation, installation guides, and related skill recommendations.
Each skill page provides copy-ready git clone commands for HTTPS, SSH, and GitHub CLI methods. You clone the skill repository from GitHub and integrate it into your Claude Code environment following the skill's README instructions.
Claude Code is an agentic coding tool that runs in the terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows through natural language commands.
The platform provides free access to browse and discover skills. Individual skills are sourced from public GitHub repositories and carry their own open-source licensing terms. Users should check each skill's repository for specific license details.
Yes, the platform supports both English and Simplified Chinese with dedicated zh/ locale paths. Skill descriptions and interface elements are culturally adapted rather than relying on raw machine translation.
Skills are community-contributed from public GitHub repositories. The platform provides AI-generated summaries and structured listings but does not currently offer independent editorial verification of skill claims. Developers should evaluate skill quality by inspecting the source repository directly.
请在官网核验
继续探索
相近任务的不同路径
这些工具以带有明确编辑理由的替代关系关联到当前产品。
Genspark.ai
Integrates top AI services with document processing and coding features, offering a broader commercial-grade AI platform compared to Agent Skills' focused skill catalog approach.
查看档案
Girikon.AI
Provides enterprise-focused AI agent development capabilities, complementing Agent Skills' community-driven, open-source skill discovery model.
查看档案
Moltbot
OpenClaw offers an alternative agent framework with its own extensibility ecosystem and open-source approach to agent skill management.
查看档案