功能描述

24 AI agents for business automation - sales outreach, lead gen, content creation, SEO, ad copy, bookkeeping, proposals, market research, business plans, tec...

使用说明 (SKILL.md)

\r \r

⚡ Digital Labour — 24 AI Agents for Business Automation\r

Name: Digital Labour
Author: resonanceenergy

\r

Your entire back-office, automated. Run any of 24 specialized AI agents through plain English. Sales outreach, lead generation, content repurposing, SEO, ad copy, bookkeeping, proposals, market research, tech docs — all with built-in QA verification.\r \r

Quick Start\r

\r Set your API URL (or use the default production endpoint):\r

export DIGITAL_LABOUR_API_URL="https://bitrage-labour-api-production.up.railway.app"\r
```\r
\r
Test the connection:\r
```bash\r
python3 {baseDir}/scripts/dl-api.py health\r
```\r
\r
Run an agent:\r
```bash\r
python3 {baseDir}/scripts/dl-api.py run support_ticket '{"ticket_text": "My order has not arrived after 2 weeks"}'\r
```\r
\r
## Available Agents (24 Total)\r
\r
### Revenue & Sales (5 agents)\r
| Agent | Command | What it does |\r
|-------|---------|-------------|\r
| **sales_outreach** | `run sales_outreach '{"company":"Stripe","role":"Head of Sales"}'` | Company research + 3-email outreach sequence |\r
| **lead_gen** | `run lead_gen '{"industry":"fintech","region":"North America"}'` | Generate qualified lead lists for any industry |\r
| **email_marketing** | `run email_marketing '{"product":"SaaS CRM","audience":"small business owners"}'` | Full email campaign with subject lines + body copy |\r
| **proposal_writer** | `run proposal_writer '{"project":"Website redesign","client_name":"Acme Corp"}'` | Professional project proposals with pricing |\r
| **ad_copy** | `run ad_copy '{"product":"AI writing tool","platform":"google"}'` | Platform-optimized ad copy (Google/Facebook/Instagram) |\r
\r
### Content & Marketing (4 agents)\r
| Agent | Command | What it does |\r
|-------|---------|-------------|\r
| **content_repurpose** | `run content_repurpose '{"content":"\x3Cblog post text>"}'` | Repurpose content into tweets, LinkedIn, newsletters |\r
| **seo_content** | `run seo_content '{"keyword":"AI automation","content_type":"blog"}'` | SEO-optimized content (blog/landing/pillar pages) |\r
| **social_media** | `run social_media '{"topic":"AI trends","platform":"linkedin","cta_goal":"drive signups"}'` | Platform-native social posts with CTAs |\r
| **press_release** | `run press_release '{"announcement":"Product launch","company":"Bitrage"}'` | PR-ready press releases |\r
\r
### Operations & Data (4 agents)\r
| Agent | Command | What it does |\r
|-------|---------|-------------|\r
| **data_entry** | `run data_entry '{"source_data":"\x3Craw data>","output_format":"CSV"}'` | Structure and clean raw data |\r
| **web_scraper** | `run web_scraper '{"source_url":"https://example.com","extraction_target":"pricing"}'` | Extract structured data from web pages |\r
| **crm_ops** | `run crm_ops '{"contact_data":"\x3Ccontact info>","action":"segment"}'` | CRM updates, segmentation, and reporting |\r
| **bookkeeping** | `run bookkeeping '{"transactions":"\x3Ctransaction data>","period":"monthly"}'` | Transaction categorization and financial reports |\r
\r
### Documents & Research (4 agents)\r
| Agent | Command | What it does |\r
|-------|---------|-------------|\r
| **doc_extract** | `run doc_extract '{"document_text":"\x3Cdoc text>","doc_type":"invoice"}'` | Extract structured data from invoices, contracts, resumes |\r
| **market_research** | `run market_research '{"topic":"electric vehicles","depth":"detailed"}'` | Market analysis at overview/detailed/comprehensive depth |\r
| **business_plan** | `run business_plan '{"business_idea":"AI tutoring app","market":"K-12 education"}'` | Full business plans with financial projections |\r
| **tech_docs** | `run tech_docs '{"code_or_api":"\x3Ccode or API spec>","doc_type":"api"}'` | API docs, READMEs, and tutorials |\r
\r
### Professional Services (3 agents)\r
| Agent | Command | What it does |\r
|-------|---------|-------------|\r
| **support_ticket** | `run support_ticket '{"ticket_text":"\x3Ccustomer issue>"}'` | Categorize, prioritize, draft reply for support tickets |\r
| **product_desc** | `run product_desc '{"product_specs":"\x3Cspecs>","tone":"luxury"}'` | Product descriptions in any tone |\r
| **resume_writer** | `run resume_writer '{"career_data":"\x3Ccareer history>","target_industry":"tech"}'` | ATS-optimized resumes for any industry |\r
\r
### Management Layer (4 agents)\r
| Agent | Command | What it does |\r
|-------|---------|-------------|\r
| **context_manager** | Internal orchestration | Maintains context across multi-agent pipelines |\r
| **qa_manager** | Internal orchestration | Quality assurance on every agent output |\r
| **production_manager** | Internal orchestration | Workflow scheduling and resource allocation |\r
| **automation_manager** | Internal orchestration | Autonomous task routing and retry logic |\r
\r
## Helper Script\r
\r
`scripts/dl-api.py` — Python script (stdlib only) for all API operations.\r
\r
**Commands:**\r
| Command | Description |\r
|---------|-------------|\r
| `health` | Check API health status |\r
| `agents` | List all available agents with input schemas |\r
| `run \x3Cagent> \x3Cjson_inputs>` | Run a specific agent with JSON inputs |\r
| `batch \x3Cjson_file>` | Run multiple agents from a JSON batch file |\r
\r
**Examples:**\r
```bash\r
# Check health\r
python3 {baseDir}/scripts/dl-api.py health\r
\r
# List all agents\r
python3 {baseDir}/scripts/dl-api.py agents\r
\r
# Generate sales outreach\r
python3 {baseDir}/scripts/dl-api.py run sales_outreach '{"company":"Tesla","role":"VP Engineering"}'\r
\r
# Write a business plan\r
python3 {baseDir}/scripts/dl-api.py run business_plan '{"business_idea":"AI-powered pet care app","market":"pet owners 25-45"}'\r
\r
# Generate SEO content\r
python3 {baseDir}/scripts/dl-api.py run seo_content '{"keyword":"remote work tools 2026","content_type":"pillar"}'\r
\r
# Run a batch of tasks\r
python3 {baseDir}/scripts/dl-api.py batch {baseDir}/examples/batch.json\r
```\r
\r
## Batch Processing\r
\r
Create a JSON file with multiple tasks:\r
```json\r
[\r
  {"agent": "sales_outreach", "inputs": {"company": "Stripe", "role": "CTO"}},\r
  {"agent": "ad_copy", "inputs": {"product": "AI CRM", "platform": "facebook"}},\r
  {"agent": "seo_content", "inputs": {"keyword": "business automation", "content_type": "blog"}}\r
]\r
```\r
\r
Run them all:\r
```bash\r
python3 {baseDir}/scripts/dl-api.py batch tasks.json\r
```\r
\r
## Multi-Agent Workflows\r
\r
Chain agents together for complex operations:\r
\r
**Lead-to-Close Pipeline:**\r
1. `lead_gen` → find prospects in target industry\r
2. `market_research` → understand their pain points\r
3. `sales_outreach` → personalized outreach sequence\r
4. `proposal_writer` → ready proposal when they respond\r
\r
**Content Engine:**\r
1. `market_research` → trending topics in your niche\r
2. `seo_content` → long-form SEO blog post\r
3. `content_repurpose` → tweets, LinkedIn, newsletter from that post\r
4. `social_media` → platform-native posts with CTAs\r
5. `ad_copy` → paid promotion copy\r
\r
**Client Onboarding:**\r
1. `doc_extract` → pull data from client documents\r
2. `data_entry` → structure into your system format\r
3. `bookkeeping` → set up financial tracking\r
4. `crm_ops` → create CRM records\r
\r
## API Details\r
\r
- **Base URL**: `https://bitrage-labour-api-production.up.railway.app`\r
- **Universal Endpoint**: `POST /v1/run`\r
- **Agent List**: `GET /agents`\r
- **Health Check**: `GET /health`\r
- **Docs**: `GET /docs`\r
- **LLM Providers**: OpenAI (GPT-4o), Anthropic (Claude), Google (Gemini), xAI (Grok)\r
- **QA**: Every output passes through QA verification before returning\r
\r
## Environment Variables\r
\r
| Variable | Required | Default | Description |\r
|----------|----------|---------|-------------|\r
| `DIGITAL_LABOUR_API_URL` | Yes | `https://bitrage-labour-api-production.up.railway.app` | API base URL |\r
| `DIGITAL_LABOUR_API_KEY` | No | *(none)* | Optional API key for authenticated access |\r
\r
## Important Notes\r
\r
- Each agent call takes 3-15 seconds (real LLM inference, not cached)\r
- All outputs include QA verification status\r
- The `provider` field is optional in all agent inputs — omit to use the server default\r
- Management agents (context_manager, qa_manager, etc.) are internal and run automatically during pipelines\r
- Batch mode processes tasks sequentially to respect rate limits\r
- All agent outputs are structured JSON with consistent schema\r

安全使用建议

This skill is functionally what it says: it forwards your agent inputs to an external Digital Labour API and returns results. Before installing, consider: - Data sensitivity: any inputs you send (documents, customer data, financials) will be transmitted to the configured BASE_URL. By default the skill uses a production third-party URL; if you do not explicitly set DIGITAL_LABOUR_API_URL you may send data to that service unintentionally. Do not use with sensitive data unless you trust and have reviewed the service's privacy and retention policies. - Missing declaration: the client will include an X-Api-Key header if DIGITAL_LABOUR_API_KEY is set, but that env var is not documented in the skill metadata — confirm expected auth methods with the publisher. - Pipeline behavior: the shipped workflows.json uses placeholders like $FROM_STEP_1.leads[0].company, but the pipeline runner performs only simple $KEY string substitution and stores prior step results as JSON strings. Expect the provided pipelines to behave incorrectly; test on non-sensitive inputs and ask the publisher for clarification or a fix. - Verification: run health/agents against a benign local or controlled endpoint first, or override DIGITAL_LABOUR_API_URL to a test server, to observe behavior. If you intend to use it in production, request from the publisher: (1) clear documentation of env vars and auth, (2) an explanation of data retention and access controls for the API, and (3) a corrected pipeline substitution mechanism if you plan to use the provided multi-step workflows.

功能分析

Type: OpenClaw Skill Name: digital-labour Version: 1.0.0 The digital-labour skill bundle provides a legitimate and well-structured interface for a business automation API hosted on Railway. The Python scripts (dl-api.py and dl-pipeline.py) use only the standard library to perform API requests and manage multi-agent workflows, with no evidence of data exfiltration, obfuscation, or malicious execution. The instructions in SKILL.md are consistent with the provided code and do not contain any prompt-injection attempts or unauthorized access requests.

能力评估

✓ Purpose & Capability

Name/description (24 business automation agents) align with the shipped code: two Python CLI scripts that call an external API to list/run agents and execute pipelines. Requiring python3 and a DIGITAL_LABOUR_API_URL is consistent with that purpose.

⚠ Instruction Scope

The SKILL.md and scripts direct the agent to send arbitrary agent inputs (potentially sensitive user data) to the configured BASE_URL. The pipeline implementation uses a simple string substitution mechanism but the provided pipelines.json expects dotted-path substitution (e.g., $FROM_STEP_1.leads[0].company). In practice the code sets FROM_STEP_n to a JSON string and substitute_vars only replaces whole $KEY tokens, so the dotted-path placeholders will not be resolved as authored — pipelines may fail or behave unexpectedly.

✓ Install Mechanism

No install spec and the shipped Python scripts use only the standard library. Nothing is downloaded or extracted during install; this is low-risk from an install-mechanism perspective.

⚠ Credentials

Declared required env var is DIGITAL_LABOUR_API_URL (a URL). The code will also read DIGITAL_LABOUR_API_KEY if present, but that env var is not declared in requires.env — a mismatch. Also, the skill defaults to a production third-party URL if the user does not set the env var, meaning data will be sent externally by default; consider this significant from a data-exfiltration/privacy perspective.

✓ Persistence & Privilege

always is false and the skill does not attempt to modify system or other skill configurations. It only reads local workflow files shipped with the skill. No elevated persistence or cross-skill config writes detected.

版本历史

v1.0.0

- Initial release of digital-labour with 24 AI agents covering business automation tasks including sales, lead generation, content creation, SEO, ad copy, bookkeeping, proposals, market research, business plans, tech docs, data entry, and more. - Multi-agent pipelines supported, featuring built-in QA verification on every output. - Powered by GPT-4o, Claude, Gemini, and Grok models. - Includes a Python stdlib-only helper script for easy API access and batch processing. - Extensive usage examples, agent documentation, and workflow suggestions provided in SKILL.md.

元数据

Slug digital-labour

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题

Digital Labour 是什么？

24 AI agents for business automation - sales outreach, lead gen, content creation, SEO, ad copy, bookkeeping, proposals, market research, business plans, tec... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 270 次。

如何安装 Digital Labour？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install digital-labour」即可一键安装，无需额外配置。

Digital Labour 是免费的吗？

是的，Digital Labour 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

Digital Labour 支持哪些平台？

Digital Labour 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 Digital Labour？

由 ResonanceEnergy（@resonanceenergy）开发并维护，当前版本 v1.0.0。

Digital Labour