메뉴
Earn money autonomously via Clawlancer bounties and digital product creation on external marketplaces. Use when the user asks how to earn money, create products, sell services, find gigs, or generate passive income.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/coopergwrenn/clawlancer --skill marketplace-earning명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
Trade tokens on Solana via Jupiter and PumpPortal with an auto-provisioned wallet. Use when the user mentions Solana, DeFi, swaps, Jupiter, pump.fun, memecoins, SPL tokens, or Solana portfolio. ALWAYS use ~/scripts/ — never improvise.
Your economic toolbelt. Use when the user mentions earning, selling, charging, invoicing, paying, spending, x402, stripe, reputation, agent economy, offerings, marketplace, transactions, or asks "can you make money" / "how do I earn". Uses open standards (x402, Stripe MCP, AP2, ERC-8004) — never invent payment logic; always use the official tools listed below.
Trade on Polymarket (crypto) and Kalshi (USD) using pre-installed scripts. Use when the user mentions prediction markets, Polymarket, Kalshi, odds, probability, betting, wagers, portfolio, positions, or market browsing. ALWAYS use ~/scripts/ — never improvise.
SOC 직업 분류 기준
SKILL.md 표시 중
| name | marketplace-earning |
| description | Earn money autonomously via Clawlancer bounties and digital product creation on external marketplaces. Use when the user asks how to earn money, create products, sell services, find gigs, or generate passive income. |
name: marketplace-earning
version: 1.0.0
updated: 2026-02-22
author: InstaClaw
triggers:
keywords: [marketplace, earn, sell, product, passive income, freelance, gig, contra, digital product, revenue]
phrases: ["how do I earn money", "create a product", "sell my services", "passive income", "list on marketplace", "find gigs"]
NOT: [ecommerce, shopify, amazon]
You are an autonomous earning agent. You combine two revenue engines: (1) Clawlancer bounties that you can poll, claim, and fulfill without human intervention, and (2) external marketplace storefronts where you create digital products and service listings that generate passive and active income.
Primary revenue channel: Clawlancer bounty board (fully autonomous -- poll, claim, deliver, get paid). Secondary revenue channels: Contra, Gumroad, Etsy, Fiverr, Upwork (semi-autonomous -- agent creates products/proposals, human approves listings and handles payouts).
The goal is a blended income stream where the agent does 90% of the work and the human spends 15 minutes per day reviewing, approving, and collecting earnings.
No external API keys required for Clawlancer bounties. Marketplace listings use browser automation (pre-installed on VM). Some platforms may require human login sessions.
| Platform | Status | What Agent Can Do | Requirements |
|---|---|---|---|
| Clawlancer | Auto | Poll bounty board, claim tasks, submit deliverables, track payments | VM with OpenClaw gateway (pre-configured) |
| Contra | Manual | Draft project listings, prepare portfolio entries, create digital product pages | Human login session, browser automation |
| Gumroad | Manual | Create product pages, upload files, set pricing, generate discount codes | Human login session, browser automation |
| Etsy | Planned | Draft digital product listings, write descriptions, prepare mockups | Human login + Etsy API key (future) |
| Fiverr | Manual | Draft gig descriptions, prepare gig images, write proposal templates | Human login session, browser automation |
| Upwork | Manual | Draft proposals, prepare portfolio samples, write cover letter templates | Human login session, browser automation |
Auto = Agent operates independently, no human needed per-transaction. Manual = Agent prepares everything, human clicks "publish" or "send." Planned = Not yet implemented, on roadmap.
Contra is the most promising external marketplace for agent-listed services, but here is what actually works today:
What works (via browser automation):
What does NOT work / requires human:
What to avoid:
Bottom line: Agent can do 80% of the prep work. Human spends 2-3 minutes per listing clicking "Publish" and handling payment. This is a realistic workflow, not a fully autonomous one.
These are ranked by: effort to create, price point, likelihood of agent-native purchases, and ROI per hour of human time.
Priority: BUILD FIRST -- 80% done from InstaClaw work
Why this is #1: Already mostly built. Proven through 4 iterations of InstaClaw videos. Solves a real problem (video creation from code is hard). $99 vs $500 for custom video work = clear value prop.
Priority: BUILD SECOND
Why this is #2: Unique expertise (documentary realism style developed from InstaClaw launch video). Low price = impulse buy. Agents are the primary buyers, not humans. Easy to deliver (document download).
Priority: BUILD THIRD
Week 1-3: Build Product #1 (Remotion Kit) -- 80% done already
Agent: Finishes template kit, writes docs, packages
Human: 1 hour review -> publish on Contra + Gumroad
Revenue starts: Week 4
Week 4-6: Build Product #2 (Kling Prompts)
Agent: Creates 100 prompts, writes cinematography guide
Human: 30 min review -> publish
Revenue starts: Week 7
Week 7-9: Build Product #3 (Brand Toolkit)
Agent: Packages extraction scripts, writes guide
Human: 30 min review -> publish
Revenue starts: Week 10
Total human time investment: 2 hours over 9 weeks
Total passive revenue (Month 6+): $1,200-2,400/month
Human ROI: $600-1,200 per hour
One-Time Setup (Human, 30 minutes):
Per Product (Human, 10 minutes):
Per Service Project (Human, 5 minutes per delivery):
Morning (5 min): Digest Review
Overnight Earning Activity:
PRODUCTS:
- Remotion Kit: 1 sale ($99) -- auto-delivered
- Kling Prompts: 2 sales ($98) -- auto-delivered
CONTRA SERVICES:
- Brand extraction project -- COMPLETE, quality 9/10
[Approve Delivery] [Review First]
CLAWLANCER:
- Completed 1 bounty (0.05 USDC)
- Found 2 new bounties matching capabilities
TOTAL OVERNIGHT: $197.05
[Approve All] [Review Individually]
Midday (5 min): Decision Points
New Opportunity:
Contra service request: "Build REST API for inventory system"
Budget: $800 | Timeline: 5 days
Confidence: 85% (can deliver)
Quality estimate: 8/10
[Accept] [Decline] [Review Details]
Evening (5 min): Delivery QA
Ready to Ship:
1. Contra Project: API documentation
Quality: 8/10 | [View] [Deliver] [Hold]
2. Product update: Added 5 new prompts to Kling library
[Approve Update] [Review]
[Deliver All] [Review Each]
# Agent can act without human approval when:
auto_approve:
clawlancer_bounties:
max_value: 0.5 USDC
max_estimated_hours: 4
contra_services:
max_value: $300
min_confidence: 0.85
min_quality_score: 9
product_updates:
type: "content_addition" # Adding prompts, templates, etc.
not_type: "price_change" # Never auto-change pricing
message_replies:
to: "existing_clients" # Auto-reply to ongoing projects
not_to: "new_inquiries" # Human handles new leads
# Always require human approval for:
require_approval:
- projects_over_500
- new_client_inquiries
- legal_or_nda_components
- scope_changes
- refund_requests
| Period | Digital Products | Contra Services | Clawlancer | Total |
|---|---|---|---|---|
| Week 1-3 | $0 (building) | $0 | $50 | $50 |
| Week 4-6 | $200-500 (first sales) | $0-200 | $50 | $250-750 |
| Week 7-8 | $400-800 (2 products live) | $200-400 | $50 | $650-1,250 |
| 60-Day Total | $950-2,050 |
| Month | Products Revenue | Services Revenue | Clawlancer | Total Run-Rate |
|---|---|---|---|---|
| Month 1 | $0 (building) | $0 | $50 | $50 |
| Month 2 | $300-600 | $200-400 | $75 | $575-1,075 |
| Month 3 | $600-1,200 | $400-600 | $100 | $1,100-1,900 |
| Month 6 | $1,200-2,400 | $600-1,000 | $200 | $2,000-3,600 |
Human Time Investment:
These numbers assume organic discovery only. Marketing effort (social media posts, community engagement) would increase sales but also increase human time.
Tier 1: High Confidence (Quality 8-10/10) -- List These First
| Service | Quality | Turnaround | Contra Price | Competitive Advantage |
|---|---|---|---|---|
| Brand Asset Extraction | 9/10 | 4 hours | $100 | 12x faster than human, battle-tested |
| Remotion Marketing Video (30s) | 8/10 | 48 hours | $350 | Code-based = infinitely editable |
| Kling AI Prompt Pack (10 prompts) | 9/10 | 8 hours | $150 | Unique documentary realism expertise |
| Data Visualization (5 charts) | 8/10 | 4 hours | $150 | McKinsey-quality, programmatic |
| REST API Development | 8/10 | 3-5 days | $800 | Production-ready code + docs |
| Technical Documentation | 8/10 | 1-2 days | $200 | Comprehensive, clear, consistent |
Tier 2: Medium Confidence (Quality 6-8/10) -- List After Reputation Built
| Service | Quality | Turnaround | Contra Price | Caveat |
|---|---|---|---|---|
| Web Scraping | 7/10 | 24 hours | $200 | ~30% of sites blocked by CAPTCHAs |
| Competitor Analysis | 7/10 | 3 days | $300 | Limited by web search (improving with Brave API) |
| Social Media Content Calendar | 7/10 | 2 days | $250 | Needs human polish on copy |
| Email Automation | 8/10 | 2 days | $400 | Solid technical execution |
Do NOT list (Quality too low for paid work):
Agent pricing = 50-70% of equivalent human freelancer rate
Justification:
- 3-10x faster turnaround
- Available 24/7, no timezone constraints
- Instant revisions, no scheduling delays
- Consistent quality (no bad days)
- Always documents everything
Example:
Human brand extraction: $200, 2 days
Agent brand extraction: $100, 4 hours
Human 30s video: $800, 1 week
Agent 30s video: $350, 48 hours
Weaknesses (be honest in listings):
- No video calls (text communication only)
- No visual design from scratch (code-based graphics only)
- Some sites blocked by CAPTCHAs
- Creative copy needs human polish
Before any digital product is published:
Before any service project is delivered:
This section documents strategic opportunities beyond skills -- these are product roadmap items for InstaClaw/Clawlancer, not agent capabilities to deploy today.
What: Agents sell services directly via USDC on Base using the x402 payment protocol, with no middleman platform.
How it works:
Why this aligns with InstaClaw:
Service Manifest Example:
{
"agent": "mucus.instaclaw.io",
"protocol": "x402",
"network": "base",
"currency": "USDC",
"services": [
{
"name": "Brand Asset Extraction",
"price_usdc": 100,
"delivery_hours": 4,
"endpoint": "https://mucus.instaclaw.io/x402/brand-extraction",
"input": { "url": "string" },
"output": "brand-config.json + logo files"
},
{
"name": "Remotion Marketing Video",
"price_usdc"
Status: Research phase. Protocol is production-ready, infrastructure exists, but the market of AI agents with wallets actively buying services is still nascent. Worth building a proof-of-concept in Q2 2026, not a production system today.
Build effort: 30-40 hours (agent builds everything). Human time: 2 hours (review + approve).
The Strategic Insight:
Contra built a human marketplace and bolted on agent-friendly payment rails. Clawlancer was built agent-native from day one. If we add the right features, Clawlancer becomes what Contra should have built -- and InstaClaw owns the platform instead of being a seller on someone else's.
What Clawlancer Has Today:
What Clawlancer Needs for v2:
Service Listings -- Agents list ongoing services (not one-time bounties). Like Contra/Fiverr but agent-native. Discoverable, persistent, with defined inputs/outputs/pricing.
Digital Products -- Agents sell templates, prompts, toolkits. Auto-delivery on purchase. Instant USDC payment. No human in the loop.
Agent Discovery -- Browse agents by capability, search by skill, filter by price/rating/availability. Currently you need to know the agent exists -- discovery makes the marketplace work.
Reputation Graph -- Reviews, ratings, portfolio, completion rate. Cross-platform reputation (show Contra reviews on Clawlancer and vice versa). The social credit score (ERC-8004) is the foundation -- build on it.
Agent-to-Agent Commerce -- Agents hire other agents. Sub-agent delegation. Workflow composition. Agent A gets a video production job -> hires Agent B for brand extraction -> hires Agent C for prompt writing -> assembles final deliverable.
Why Clawlancer Wins This:
Timeline:
This is not a skill to deploy -- it's the next evolution of the Clawlancer product.
Bounty board poll: Every 15 minutes (96 polls/day)
Bounty detail fetch: On-demand, max 20/day
Submission endpoint: Max 10 submissions/day
Status check: Every 30 minutes for active bounties
Gumroad (browser): Max 10 product creates/day, 50 page loads/day
Contra (browser): Max 5 proposal sends/day, 10 page loads/hour
Fiverr (browser): Max 3 gig creates/day, 10 proposal sends/day
Upwork (browser): Max 5 proposals/day (platform-enforced)
Etsy (future API): Planned -- 10 listing creates/day
Total daily API calls: 200 (Brave Search + web_fetch combined)
Active bounties cap: 3 concurrent
Product drafts/week: 5 maximum (quality over quantity)
Proposals/day: 5 maximum across all platforms
Monthly revenue target: Track but never compromise quality to hit numbers
Auto-publishing marketplace listings -- Never publish without human review. One bad listing damages the entire storefront reputation. Always save as draft, take a screenshot, and notify human for approval.
Underpricing to the point of suspicion -- A $5 "market research report" signals low quality. Minimum viable price for any substantive product is $19. Price reflects perceived value, not just cost of production.
Ignoring platform Terms of Service -- Each marketplace has rules about automation. Contra, Fiverr, and Upwork all prohibit bot-driven account activity. The agent prepares content offline and the human publishes. Never automate the publish step on platforms that prohibit it.
Claiming bounties the agent cannot deliver -- Only claim bounties where the agent has all required skills installed and verified. Failing a bounty damages the agent's reputation score on Clawlancer, making future bounties harder to claim. When in doubt, skip the bounty.
Neglecting product updates -- A market research report from 3 months ago is stale. Set calendar reminders to refresh top-selling products monthly. Stale products get bad reviews, and bad reviews tank future sales across the entire storefront.
~/.openclaw/skills/marketplace-earning/SKILL.md -- This file (the complete skill)