Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Polymarket Brain

v1.0.0

Automates geopolitical and macroeconomic news analysis, matches expert insights to Polymarket odds, and posts actionable trade recommendations to Discord.

0· 190·0 current·0 all-time
byDimas Chandra@dimaschand29

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for dimaschand29/polymarket-brain.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Polymarket Brain" (dimaschand29/polymarket-brain) from ClawHub.
Skill page: https://clawhub.ai/dimaschand29/polymarket-brain
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install polymarket-brain

ClawHub CLI

Package manager switcher

npx clawhub@latest install polymarket-brain
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
High-level purpose (news → expert analysis → Polymarket match → Discord) matches the code and artifacts: news fetch, analyzer, market mapping, Discord payloads. However the SKILL.md and other docs instruct copying the entire ~/.browseros/skills tree and editing orchestrator files; the skill also claims to write into BrowserOS core memory and various user-profile paths. Copying or modifying other skills' directories and core memory is not necessary for a simple market-analysis pipeline and is disproportionate to the stated purpose.
!
Instruction Scope
Runtime instructions include editing files and hardcoded webhook URLs, running scripts that assume other components (cnbc-geopolitics-fetcher) exist, and explicit copy commands for the user's whole skills directory. The skill also documents persistence and commands (VERIFY_INSTALLATION.bat, AFTER_RESTART_CHECK.bat, RUN_ORCHESTRATOR.bat) and claims to update BrowserOS 'core memory' — these are scope expansions beyond analysing and posting market recommendations. Hardcoded webhooks (full tokens visible in several files) mean the skill will (or can) post to external Discord channels without additional credentials.
Install Mechanism
No formal install spec is declared (instruction-only), which is lower risk for automatic installs but surprising given there are many code files bundled. There is no remote download/execute step in the manifest, but the presence of many executable scripts and .bat files (and claims they persist across restarts) means manual execution will write to disk and potentially modify environment. The absence of an install step combined with many included files is an inconsistency worth attention.
!
Credentials
The skill requires no declared env vars or credentials, but multiple files embed full Discord webhook URLs (sensitive tokens). The SKILL.md instructs editing orchestrator files to set webhooks (or uses the ones already present). It also references other skills' history files and suggests copying the entire skills folder—this gives access to other skills' files and could expose unrelated secrets. No Polymarket API credentials are requested or documented despite matching to markets; the skill appears to rely on hardcoded market URLs/odds rather than authenticated API access.
!
Persistence & Privilege
The documentation repeatedly promises multi-layer persistence (local files, BrowserOS core memory, session backups) and includes scripts to verify persistence after restart. It also recommends copying the entire skills directory and writing verification/launcher .bat files into user paths. While persisting its own files is reasonable, the explicit instructions to copy or modify the user's skills directory and core memory are elevated privileges relative to a news-analysis skill and could overwrite or read other skills' config/memory.
What to consider before installing
What to consider before installing or running this skill: - Do not run bundled scripts or .bat files blindly. Review the orchestrator (polymarket_brain_orchestrator.py and scripts/*) to see exactly what files and paths they read/write and what network endpoints they call. - The skill contains full Discord webhook URLs (tokens) in multiple files. Those allow posting to third-party channels immediately; confirm you control the target webhook or replace it with a webhook you control before running. - The SKILL.md instructs copying your entire skills directory and writing to BrowserOS core memory—both can expose or overwrite other skills and their secrets. Avoid mass-copy steps; perform any migration manually and with backups. - Polymarket interactions appear hardcoded to specific market URLs and odds; there is no documented credential handling for Polymarket. If you expect real API calls, verify where and how API keys would be stored and used. - If you want to try it safely: run it in an isolated sandbox/VM or on a throwaway account, remove or replace webhook URLs, and search the code for any network calls beyond Discord and Polymarket (look for requests.post/urllib/ socket usage). Also inspect any code that writes to %APPDATA%/BrowserOS or other user profile paths. If you want, I can (a) list specific files and lines that write to user/home paths or contain the webhook tokens, (b) search the code for outbound network endpoints beyond Discord/Polymarket, or (c) walk you through a safe checklist to run it in a sandbox. Confidence in this assessment is medium because many behaviors are plausible for a local orchestration tool, but the filesystem and persistence instructions are broader than expected for the stated purpose.

Like a lobster shell, security has layers — review code before you run it.

latestvk97fyk1dctt232nh71add8swh183488v
190downloads
0stars
1versions
Updated 22h ago
v1.0.0
MIT-0

Polymarket-Brain v1.1 - Complete Skill Documentation

🎯 Purpose

Automated geopolitical and macroeconomic analysis pipeline that:

  1. Fetches breaking news from CNBC World Politics
  2. Routes to expert analysts (geopolitics-expert / the-fed-agent)
  3. Matches analyses to Polymarket prediction markets
  4. Delivers actionable trading recommendations to Discord

End Goal: Identify market mispricing opportunities where expert probability diverges from market odds.


🏗️ Architecture (4 Phases)

Phase 1: CNBC News Fetching

Component: cnbc-geopolitics-fetcher

# Execution
cd cnbc-geopolitics-fetcher
python scripts/fetch_cnbc_geopolitics.py --config references/config.md --count 5

Behavior:

  • Fetches exactly 5 articles from CNBC World Politics RSS
  • Deduplicates against references/sent_urls.txt history
  • Posts NEW articles individually to Discord (not batched)
  • Exit code 0 always (even when no new articles)

Webhook: https://discord.com/api/webhooks/1482043765471445333/...

History Management:

  • Tracks all sent URLs in sent_urls.txt
  • Skips duplicates forever
  • Growth = fewer posts over time (expected, not broken)

No New Articles Behavior:

  • Does NOT post to Discord
  • Returns exit code 0
  • Orchestrator skips Phases 2-4 (no analysis, no market matching, no Discord posts)
  • Prints console notification only (not sent to Discord)
  • Clean exit with clear status message

Phase 2: Expert Analysis

Classification Logic:

News ContentClassificationExpert Skill
Military strikes, troop movements, Iran, Russia, Middle East, oil, warGeopoliticsgeopolitics-expert
Fed decisions, inflation, employment, Treasury yields, rates, currencyMacroeconomicsthe-fed-agent
Mixed (war + markets, oil + Fed)BothRun both skills

Expert Models:

geopolitics-expert

Frameworks:

  • IRGCistan: Military-dominated state analysis
  • Hormuz Siege: Blockade impact assessment
  • Five Pathways: Conflict termination probabilities

Outputs:

  • Conflict duration probability
  • Commodity impacts (oil, gold, USD)
  • Trading opportunities

the-fed-agent

Frameworks:

  • Dot Plot Analysis: Fed stance changes
  • Oil Stagflation: Energy shock impact
  • FX Strategies: EUR/USD positioning

Outputs:

  • Rate hike probability
  • Stagflation risk assessment
  • Currency recommendations

Phase 3: Market Matching

8 Monitored Polymarket Markets:

#Market NameResolution DateExpert Probability
1Will Iranian regime fall by June 30?Jun 30, 202615%
2US-Iran ceasefire by December 31Dec 31, 202655%
3Iran conflict ends by December 31Dec 31, 202660%
4US forces enter Iran by December 31Dec 31, 202635%
5Iran leadership change by December 31Dec 31, 202685%
6Fed decision in MarchMar 202695%
7Will crude oil CL hit $100+ by end of MarchMar 202690%
8US recession by end of 2026End 202640%

Matching Process:

  1. Extract market odds from Polymarket (hardcoded in v1.1)
  2. Apply expert probability assessment from Phase 2
  3. Calculate edge (market vs expert mispricing)
  4. Generate recommendation (Strong Yes/No, Lean, Fair Value)

Recommendation Logic:

Market OddsExpert ProbGapRecommendation
< 30% YesExpert > 50%20%+Strong Yes
> 70% YesExpert < 50%20%+Strong No
SimilarSimilar< 10%Fair Value ⚖️
Expert > 70%Market < 70%-Underpricing
Expert < 40%Market > 60%-Overpricing

Phase 4: Discord Dispatch

Message Format:

Header:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🧠 POLYMARKET-BRAIN WORKFLOW ACTIVATED
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Time: YYYY-MM-DD HH:MM:SS
Version: 1.1
Markets Analyzing: X

Each Market (sent individually, 1.2s delay):

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🧠 POLYMARKET-BRAIN ANALYSIS: MARKET N/X
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

📌 MARKET: [Market Name]

📅 Resolution Date: [Date]
📊 Market Odds: XX% Yes
🎯 Expert Probability: XX% Yes
💡 Recommendation: [Emoji] [Text]

📝 REASONING:
[Detailed reasoning from expert analysis]

🔗 Link: [Polymarket URL]

📈 Edge: XX% market mispricing

Analyzed by: [expert] | Confidence: HIGH

Summary:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🧠 POLYMARKET-BRAIN WORKFLOW COMPLETE
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

📊 Summary:
• Total Markets Analyzed: X
• Strong Buys (Yes): X
• Strong Sells (No): X
• Expert Sources: [list]
• Data Source: CNBC World Politics

⏱️ Workflow Time: [timestamp]
🔄 Next Update: Check back for fresh analysis

Disclaimer: This is automated analysis. Not financial advice.

Rate Limit: 1.2 seconds between ALL messages (prevents Discord 429 rate limiting)

Webhook: https://discord.com/api/webhooks/1483478506070474922/...


🔧 Installation & Migration

Fresh Environment Setup

Step 1: Copy Skills

# Windows
xcopy /E /I "C:\Users\Legion 5i Pro\.browseros\skills" "D:\new-env\.browseros\skills"

# Linux/Mac
cp -r ~/.browseros/skills /new/path/.browseros/skills

Step 2: Verify Python Dependencies

pip install requests crawl4ai python-dateutil
pip list | findstr "requests crawl4ai"

Step 3: Configure Webhooks

CNBC Fetcher: Edit cnbc-geopolitics-fetcher/references/config.md:

DISCORD_WEBHOOK=https://discord.com/api/webhooks/1482043765471445333/YOUR_KEY_HERE

Polymarket Brain: Edit polymarket_brain_orchestrator.py (lines 15-16):

PHASE1_DISCORD_WEBHOOK = "https://discord.com/api/webhooks/1482043765471445333/YOUR_KEY"
PHASE4_DISCORD_WEBHOOK = "https://discord.com/api/webhooks/1483478506070474922/YOUR_KEY"

Step 4: Test Run

cd polymarket-brain
python polymarket_brain_orchestrator.py

Expected: 5 articles posted, 5-8 markets analyzed, all sent to Discord.


🚨 Troubleshooting Guide

Issue: 403 Forbidden on Discord Post

Cause: Missing User-Agent header in urllib requests

Fix:

req = urllib.request.Request(webhook_url, data=data, headers={
    'User-Agent': 'Polymarket-Brain/1.1',
    'Content-Type': 'application/json'
})

Issue: 400 Bad Request

Cause: Message exceeds Discord 2000 character limit

Fix: Shorten separator from "━" * 50 to fixed 40 chars

Issue: No Articles Posted to Discord

Cause: All URLs already in sent_urls.txt history

Fix: This is EXPECTED behavior. History growth = fewer new articles. For testing: type nul > references/sent_urls.txt

Issue: Invalid Polymarket URL (404)

Cause: Double underscores in URL (__) instead of hyphens

Fix:

# Wrong: https://polymarket.com/event/will-crude-oil-cl-hit__-by-end-of-march
# Right: https://polymarket.com/event/will-crude-oil-cl-hit-100-by-end-of-march

Issue: Placeholder Reasoning in Messages

Cause: Hardcoded text instead of expert analysis

Fix: Use actual reasoning field from expert analysis output

Issue: Rate Limiting (429 Too Many Requests)

Cause: Rapid sequential posts without delay

Fix: Add time.sleep(1.2) between ALL Discord messages

Issue: PowerShell Variable Parsing Errors

Cause: PowerShell one-liners with %VARIABLE% syntax

Fix: Use batch files (.bat) instead of PowerShell for Windows

Issue: Path with Spaces Fails

Cause: "Legion 5i Pro" breaks command parsing

Fix: Use short name Legion~1 or full quotes:

cd "C:\Users\Legion 5i Pro\.browseros\skills\polymarket-brain"

📁 File Structure

polymarket-brain/
├── polymarket_brain_orchestrator.py    # Main orchestrator (edit webhooks here)
├── SKILL.md                            # This documentation
├── README.md                           # Quick start guide
├── VERSION                             # v1.1
├── output/                             # Analysis JSON outputs (auto-created)
│   ├── analysis_input_1.json
│   ├── analysis_input_2.json
│   └── ...
└── logs/                               # Execution logs (auto-created)

cnbc-geopolitics-fetcher/
├── scripts/
│   └── fetch_cnbc_geopolitics.py       # News fetcher script
├── references/
│   ├── config.md                       # Webhook config (LOCKED - edit here)
│   └── sent_urls.txt                   # URL history (clear for test)
├── SKILL.md                            # Detailed fetcher docs
└── LOCKED.md                           # Protection marker (READ BEFORE EDIT)

🧪 Testing Commands

Production Run

cd C:\Users\Legion 5i Pro\.browseros\skills\polymarket-brain
python polymarket_brain_orchestrator.py

Fresh Test (Clear History)

type nul > C:\Users\Legion 5i Pro\.browseros\skills\cnbc-geopolitics-fetcher\references\sent_urls.txt
cd C:\Users\Legion 5i Pro\.browseros\skills\polymarket-brain
python polymarket_brain_orchestrator.py

Verify Webhooks

import requests
webhook = "https://discord.com/api/webhooks/1482043765471445333/YOUR_KEY"
response = requests.post(webhook, json={"content": "test"})
print(response.status_code)  # Should be 204

📊 Expert Probability Reference

Geopolitics-Expert Mappings

MarketExpert ProbReasoning Framework
Iran regime fall15%IRGC institutional depth, loyalty purges
US-Iran ceasefire55%Trump unilateral stance, forever war risk
Iran conflict ends60%Five Pathways, 35% forever war
US forces enter Iran35%Ground invasion unlikely, air sufficient
Iran leadership change85%ALREADY OCCURRED (Mojtaba appointed)
Oil $100+ March90%Hormuz closure 40%, oil $140-160 risk

The-Fed-Agent Mappings

MarketExpert ProbReasoning Framework
Fed decision March95%Fed HOLD at 95% per CME FedWatch
US recession 202640%Stagflationary pressure, oil shocks

🔐 Security & Best Practices

Webhook Security

  • Never commit webhook URLs to version control
  • Use environment variables in production:
    import os
    PHASE1_WEBHOOK = os.environ.get('PHASE1_DISCORD_WEBHOOK')
    

Rate Limiting

  • Always add 1.2s delay between Discord posts
  • Monitor for 429 responses
  • Implement exponential backoff if needed

History Management

  • Never delete sent_urls.txt in production
  • Only clear for testing purposes
  • Backup before clearing: copy sent_urls.txt sent_urls.txt.bak

Error Handling

  • Always exit 0 even when no new articles
  • Log all errors to logs/ directory
  • Notify user clearly when workflow stops

🔄 Version History

VersionDateChanges
v1.12026-03-19No-new-news fix: Skip Phases 2-4 when no new articles, console notification only (no Discord post), PROD/TEST mode via env var
v1.12026-03-18Production deploy, migration docs, troubleshooting guide, expert reasoning populated
v1.02026-03-18Initial orchestration, all 4 phases working, webhooks fixed, URLs valid

📞 Quick Reference

Skill Location: C:\Users\Legion 5i Pro\.browseros\skills\polymarket-brain\

Run Command:

cd C:\Users\Legion 5i Pro\.browseros\skills\polymarket-brain
python polymarket_brain_orchestrator.py

Webhooks:

  • Phase 1 (News): 1482043765471445333
  • Phase 4 (Analysis): 1483478506070474922

History File: ..\cnbc-geopolitics-fetcher\references\sent_urls.txt

Test Mode: Clear history file before run


🎓 Learning Path for New Environments

  1. Read this SKILL.md completely (you're here now ✅)
  2. Read README.md for quick start
  3. Copy skills folder to new environment
  4. Configure webhooks in both config files
  5. Install dependencies: pip install requests crawl4ai
  6. Test run with history cleared
  7. Verify Discord receives all messages
  8. Deploy to production (stop clearing history)

Time Estimate: 15-30 minutes for full setup


💾 Backup & Restore

Backup Command

xcopy /E /I "C:\Users\Legion 5i Pro\.browseros\skills\polymarket-brain" "D:\backup\polymarket-brain"
xcopy /E /I "C:\Users\Legion 5i Pro\.browseros\skills\cnbc-geopolitics-fetcher" "D:\backup\cnbc-geopolitics-fetcher"

Restore Command

xcopy /E /I "D:\backup\polymarket-brain" "C:\Users\Legion 5i Pro\.browseros\skills\polymarket-brain"
xcopy /E /I "D:\backup\cnbc-geopolitics-fetcher" "C:\Users\Legion 5i Pro\.browseros\skills\cnbc-geopolitics-fetcher"

🎯 Success Criteria

✅ Phase 1: 5 articles fetched, posted to Discord ✅ Phase 2: Expert analyses generated (geopolitics/fed) ✅ Phase 3: Markets matched with recommendations ✅ Phase 4: All messages sent to Discord (no 403/400 errors) ✅ URLs: All Polymarket links valid (no 404) ✅ Reasoning: Actual expert analysis (no placeholders) ✅ Rate Limiting: No 429 errors (1.2s delays working)

If all ✅, deployment successful.

Comments

Loading comments...