← 返回 Skills 市场
169
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install eu-ai-compliance
功能描述
EU AI Act risk classification, Article 12 compliance logging, and self-assessment reports — deadline August 2, 2026
使用说明 (SKILL.md)
EU AI Act Compliance
Free risk classification + paid full assessments. EU AI Act enforcement deadline: August 2, 2026. Fines: EUR 35M or 7% global turnover.
Base URL
https://soul.sputnikx.xyz
Endpoints
Risk Classification (FREE)
curl "https://soul.sputnikx.xyz/soul/compliance/risk-classification?description=facial+recognition+for+hiring"
Returns: { risk_level, reasoning, annex_iii_matches, disclaimer } — instant HIGH/LIMITED/MINIMAL classification.
Article 12 Mapping (FREE)
curl https://soul.sputnikx.xyz/soul/compliance/mapping
Returns: Full Article 12 → data field mapping for compliance logging requirements.
Self-Assessment ($1.00 x402 USDC)
curl https://soul.sputnikx.xyz/soul/compliance/self-assessment/{agent_id}
Complete EU AI Act self-assessment report with recommendations.
Annex IV Technical Documentation ($1.00 x402 USDC)
curl https://soul.sputnikx.xyz/soul/compliance/annex-iv/{agent_id}
Annex V EU Declaration ($0.50 x402 USDC)
curl https://soul.sputnikx.xyz/soul/compliance/annex-v
Full Compliance Report ($2.00 x402 USDC)
curl https://soul.sputnikx.xyz/soul/compliance/full-report
Complete bundle: risk classification + self-assessment + Annex IV + Annex V.
Web Interface
Free compliance checker: https://soulledger.sputnikx.xyz/compliance
Infrastructure
- Hash-chain logging (SHA-256, append-only)
- Merkle root anchoring on Base chain
- Behavioral drift detection
- Runtime monitoring (not one-time reports)
When to use this skill
- Check if your AI system is EU AI Act compliant
- Generate compliance documentation before the August 2026 deadline
- Implement Article 12 automatic logging
- Get risk classification for any AI system description
安全使用建议
This skill is an instruction-only connector to an external compliance service; that makes it low local risk but means you are trusting a third party with any text you send. Before installing or using it: 1) Verify the provider and website (the SKILL.md lists two related but different domains). 2) Do not send sensitive or proprietary system details until you confirm the service's privacy policy and data retention practices. 3) Clarify the payment flow and costs (the pricing string in SKILL.md is ambiguous: 'x402 USDC'); avoid sending funds until you understand how payments are handled. 4) If you need cryptographic proofs (Merkle root anchoring on Base chain), ask for verifiable examples and public anchors you can independently check. 5) For production use, run initial tests with non-sensitive sample data.
功能分析
Type: OpenClaw Skill
Name: eu-ai-compliance
Version: 1.0.0
The skill bundle provides a documentation-based interface for an external EU AI Act compliance service hosted at soul.sputnikx.xyz. It contains no executable code, requests no environment variables or sensitive permissions, and lacks any indicators of data exfiltration, persistence, or malicious prompt injection. The behavior is entirely consistent with its stated purpose of providing risk classification and compliance reporting.
能力评估
Purpose & Capability
Name/description match the SKILL.md: it provides risk classification, Article 12 mapping, and paid reports via HTTP endpoints. The skill requests no local binaries, env vars, or installs — which is consistent for an instruction-only HTTP API integration. Minor inconsistency: the SKILL.md metadata homepage (soulledger.sputnikx.xyz) differs from the base API (soul.sputnikx.xyz) and the ownership/hosting is 'unknown', so verify provider identity before relying on results.
Instruction Scope
Instructions only tell the agent to curl specific API endpoints on the listed domain — there are no commands to read local files or other system state. However, the skill gives no guidance about what 'agent_id' is or about authentication/payment flows. Also: any data you send (descriptions, system details) will go to an external service — avoid sending secrets or proprietary data until you've validated the provider and privacy terms.
Install Mechanism
No install spec or code is present (instruction-only). This minimizes on-disk risk because nothing is downloaded or executed locally by the skill itself.
Credentials
No environment variables, credentials, or config paths are requested. That is proportionate for a client that simply calls a public web API. Note: the SKILL.md mentions payments and on-chain anchoring but does not request wallet keys — payment likely occurs externally; confirm how payments are processed before sending funds.
Persistence & Privilege
always is false and model invocation is allowed (default). The skill does not request persistent privileges, nor does it modify system or other-skill configuration — this is appropriate for the described function.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install eu-ai-compliance - 安装完成后，直接呼叫该 Skill 的名称或使用
/eu-ai-compliance触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
- Initial release of EU AI Act compliance toolkit.
- Provides instant risk classification for any AI system.
- Offers Article 12 compliance logging and mapping.
- Supports paid self-assessments, Annex IV technical documentation, Annex V declarations, and full compliance reports.
- Features advanced infrastructure: hash-chain logging, Merkle root anchoring, behavioral drift detection, and runtime monitoring.
- Free compliance checker web interface available.
元数据
常见问题
Eu Ai Compliance 是什么？
EU AI Act risk classification, Article 12 compliance logging, and self-assessment reports — deadline August 2, 2026. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 169 次。
如何安装 Eu Ai Compliance？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install eu-ai-compliance」即可一键安装，无需额外配置。
Eu Ai Compliance 是免费的吗？
是的，Eu Ai Compliance 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Eu Ai Compliance 支持哪些平台？
Eu Ai Compliance 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Eu Ai Compliance？
由 Orion（@drivenbymyai-max）开发并维护，当前版本 v1.0.0。
推荐 Skills