← 返回 Skills 市场
872
总下载
2
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install hiring
功能描述
Hire humans or AI agents for tasks you cannot do yourself, with platform selection, vetting, contracts, and payment management.
使用说明 (SKILL.md)
Platform Selection
| Task Type | Platform | Why |
|---|---|---|
| Development, design, writing | Upwork, Freelancer.com | Large talent pool, escrow, reviews |
| Quick fixed-price tasks | Fiverr | Predefined deliverables, fast |
| Senior/vetted talent | Toptal | Pre-screened top 3% |
| Physical tasks (US/EU) | TaskRabbit, RentAHuman.ai | Local presence, verification |
| Design contests | 99designs | Multiple concepts, competition |
For API integration details, see platforms.md.
Hiring Checklist
- Define scope precisely — Deliverables, timeline, budget, success criteria
- Write compelling job post — Include context, requirements, what success looks like
- Screen candidates — Portfolio quality, relevant experience, communication, reviews
- Verify before hiring — Test task or paid trial for significant engagements
- Negotiate rates — Know market rates; prefer fixed-price or milestones over hourly
- Contract before work — IP assignment, NDA, payment terms, termination clause
- Structured onboarding — Access credentials, project brief, communication channels
- Milestone payments — Never 100% upfront; tie payments to deliverables
- Track and document — Log hours, deliverables, feedback for future reference
Red Flags
- Requests full payment upfront
- Portfolio inconsistent with claimed experience
- Vague answers to specific questions
- Sudden unavailability after deposit
- "Senior" applies, different person delivers
- Copied/AI-generated portfolio pieces
- Refuses video call or screen share
- Pushes to move off-platform immediately
Rate Benchmarks
Check current market rates at platforms.md. General guidance:
- Below market = quality or availability issues likely
- 20-30% above market = acceptable for proven performers
- Get 3+ quotes before committing on large projects
Contracts
For contract templates and payment structures, see contracts.md.
Legal Essentials
Before engaging contractors, review legal.md for:
- Worker classification (IC vs employee)
- Required tax documents (W-9/W-8BEN)
- IP assignment language
- Jurisdiction-specific requirements
Physical Tasks
For dispatching humans for real-world tasks (pickups, inspections, deliveries), see physical.md.
Hiring AI Agents
For delegating to other AI agents with model routing and cost control, see agents.md.
Contractor Management
Track per contractor:
- Skills, rates, timezone, languages
- Past projects and performance
- Availability windows
- Communication preferences
- Reliability score (1-5)
After each engagement:
- Document what worked/didn't
- Update reliability score
- Note for future matching ("great for React, slow on mobile")
安全使用建议
This skill appears to be a coherent hiring playbook and not malicious, but it assumes your agent may be connected to external platforms and payment methods. Before installing or enabling it: (1) do not provision broad payment credentials or full-control API keys to the agent—use limited-scope or test accounts; (2) configure hard budget caps and require human approval for payments above a small threshold; (3) test with low-risk/sample tasks first (no real money or sensitive data); (4) review the legal.md and contracts.md guidance and ensure tax/worker-classification compliance for your jurisdiction; (5) if you plan to enable autonomous dispatching of physical tasks, verify the reputation and insurance of the referenced platforms (e.g., RentAHuman.ai) and require proof-of-completion mechanisms (GPS/photo/video); (6) enable auditing/logging so you can review any actions the agent takes. These steps will reduce the main operational risks the skill implies.
功能分析
Type: OpenClaw Skill
Name: hiring
Version: 1.0.0
The OpenClaw AgentSkills bundle describes a legitimate 'Hiring' skill, providing comprehensive documentation for an AI agent to manage the hiring of humans and other AI agents. While the skill involves interactions with external APIs (e.g., Upwork, Freelancer.com, RentAHuman.ai) and handling potentially sensitive data (e.g., GPS-tagged photos, authentication tokens), these capabilities are inherent to its stated purpose. The documentation, including `SKILL.md`, `legal.md`, and `physical.md`, consistently promotes responsible practices, risk mitigation, and legal compliance, with no evidence of intentional harmful behavior, data exfiltration to unauthorized endpoints, or malicious prompt injection.
能力评估
Purpose & Capability
The name and description (hiring humans or AI agents, platform selection, vetting, contracts, payments) match the included documents (platforms.md, contracts.md, legal.md, physical.md, agents.md). There are no unexpected environment variables, binaries, or install steps requested that would contradict the stated purpose.
Instruction Scope
SKILL.md and the included files stay on-topic (platform selection, vetting, contract templates, agent orchestration, physical task verification). They instruct the agent to route tasks to other agents, make platform API calls, dispatch humans for physical tasks, and require logging/audit trails. The instructions do not instruct the agent to read unrelated system files or hidden endpoints, but they do implicitly assume the agent may have access to platform/payment APIs and authorization at runtime (e.g., dispatching, escrow/stablecoin payments).
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing will be written to disk or downloaded by the skill itself.
Credentials
The skill declares no required environment variables or credentials, which is consistent with an advisory/instruction-only skill. However the content references many third-party APIs (Upwork, Freelancer, RentAHuman.ai, stablecoins, etc.) that would require credentials to actually integrate. Users should not assume the skill will function autonomously without providing external credentials at runtime; any credentials supplied later should be scoped and limited.
Persistence & Privilege
The skill is not marked always:true and is user-invocable. Autonomous invocation defaults are unchanged. There is some operational risk if the agent that uses this skill is granted live payment/dispatch authority (the skill's instructions explicitly mention budget caps and 'agent should be authorized to dispatch immediately' for emergencies). That risk stems from agent permissions and credentialing, not from the skill itself.
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 hiring - 安装完成后，直接呼叫该 Skill 的名称或使用
/hiring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
元数据
常见问题
Hiring 是什么？
Hire humans or AI agents for tasks you cannot do yourself, with platform selection, vetting, contracts, and payment management. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 872 次。
如何安装 Hiring？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install hiring」即可一键安装，无需额外配置。
Hiring 是免费的吗？
是的，Hiring 完全免费（开源免费），可自由下载、安装和使用。
Hiring 支持哪些平台？
Hiring 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 Hiring？
由 Iván（@ivangdavila）开发并维护，当前版本 v1.0.0。
推荐 Skills