agentleFS
agent20usdagent20usd/claude-skills/polymarket-analyzer/SKILL.md

polymarket-analyzer

Skill1 starsChanged 8 months agoMITView on GitHub

Analyze Polymarket prediction markets AND 5-min crypto markets using technical analysis. Uses Binance price data, RSI/EMA/momentum indicators, news research, and edge detection to find profitable trades.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. mjunaidca/polymarket-skills★ 91same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use Binance 1-minute klines for real-time price data:

    line 33
  2. Distinctive:

    Wait 90 seconds into the window before analyzing (need price action data)

    line 60
  3. Distinctive:

    Use market orders for instant fills (limit orders can freeze cash)

    line 61
  4. Distinctive:

    Use web search to find:

    line 81
  5. Distinctive:

    When do they happen relative to resolution?

    line 96
  6. Distinctive:

    When an outcome has swept ALL precursor events → market underprices by 2-5c.

    line 126
  7. Distinctive:

    Place bets BEFORE data releases, not after.

    line 135
  8. Distinctive:

    If the thesis breaks, sell immediately.

    line 141

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Polymarket Analyzer v2: Prediction Markets + Crypto TA

You are an expert prediction market analyst and crypto trader. You analyze both traditional Polymarket events AND 5-minute crypto up/down markets using technical analysis signals.

Part 1: 5-Minute Crypto Trading (NEW in v2)

How It Works

Polymarket has 5-minute up/down markets for BTC, ETH, SOL, and XRP. A new window opens every 5 minutes. You predict whether the price goes UP or DOWN in the next window.

CRITICAL RULE: NEVER BET BLIND

$30 → $7 in 3 rounds betting without analysis. 50/50 + fees = guaranteed loss over time. ONLY bet when technical analysis score >= 3 (strong signal alignment).

Market API Pattern

GET https://gamma-api.polymarket.com/events?slug={coin}-updown-5m-{unix_timestamp}
  • {coin}: btc, eth, sol, xrp
  • {unix_timestamp}: window start, rounded to 300-second boundary: (now // 300) * 300
  • Token ID parsing: clobTokenIds returns as JSON string — must json.loads() it
  • Token index 0 = UP, index 1 = DOWN

Technical Analysis Signals

Use Binance 1-minute klines for real-time price data:

GET https://api.binance.com/api/v3/klines?symbol=BTCUSDT&interval=1m&limit=15

Score each indicator (+1 or -1). Sum for final score:

#IndicatorBullish (+1)Bearish (-1)
15-min Momentum> +0.05%< -0.05%
215-min Trend> +0.1%< -0.1%
3EMA5 vs EMA10EMA5 > EMA10EMA5 < EMA10
4RSI (14-period)< 30 (oversold)> 70 (overbought)
5Volume Spike> 1.5x avg (confirms trend)> 1.5x avg (confirms trend)
6Last CandleGreen (close > open)Red (close < open)

5-min momentum is the strongest signal — weight it x2 in scoring.

Decision Rules

  • Score >= 3: BET (strong signal alignment)
  • Score >= 4: HIGH CONFIDENCE — bet 15% of balance @ 52c
  • Score 3: MEDIUM CONFIDENCE — bet 10% of balance @ 50c
  • Score < 3: SKIP — patience = profit
  • Max exposure per window: 50% of balance across all coins
  • Max per coin: 20% of balance

Timing

  • Wait 90 seconds into the window before analyzing (need price action data)
  • Use market orders for instant fills (limit orders can freeze cash)
  • Fallback to aggressive limit orders if market orders fail

Recovery Proof

Using this TA system: $7.32 → $47.49 recovery. Technical analysis works on 5-min timeframes.


Part 2: Traditional Prediction Market Analysis

Analysis Framework

Step 1: Gather Market Data

For each market, collect:

  • Current price (= implied probability)
  • Volume and liquidity
  • Resolution date and criteria
  • Order book depth

Step 2: Research the Event

Use web search to find:

  • Latest news about the event
  • Historical base rates for similar events
  • Expert opinions and forecasts
  • Key catalysts (upcoming dates, decisions, announcements)

Step 3: Estimate Fair Probability

Use these techniques:

Base Rate Analysis

  • What % of similar events historically resolved YES?
  • Example: "PGA Best Picture winner wins Oscar ~87% of the time"

Catalyst Mapping

  • What events could change the probability?
  • When do they happen relative to resolution?
  • Calendar arbitrage: Does a catalyst happen AFTER the resolution date?

News Integration

  • Has something already happened that the market hasn't priced in?
  • Are there breaking developments?

Term Structure

  • Compare prices across different time horizons
  • Near-term markets should price differently than far-dated ones

Step 4: Calculate Edge

Edge = Your Fair Price - Market Price
  • Edge > 5 cents: Worth considering
  • Edge > 10 cents: Strong opportunity
  • Edge > 15 cents: High conviction trade

Step 5: Risk Assessment

Rate confidence: LOW / MEDIUM / HIGH Consider:

  • How certain are you in your fair price estimate?
  • What could go wrong? (tail risks)
  • Is the market thin? (slippage risk)
  • How close to resolution? (time value)

Proven Strategies

1. Near-Lock Categories When an outcome has swept ALL precursor events → market underprices by 2-5c.

2. Calendar Arbitrage Catalyst event AFTER resolution date → market overprices YES. Buy NO.

3. Already-Resolved Events Markets sometimes lag behind real-world results. Free money if you confirm the outcome.

4. CPI/Economic Data Plays Place bets BEFORE data releases, not after. Consensus expectations create predictable reactions.

5. Near-Expiry Impossibilities Events that clearly cannot happen in remaining time. Buy NO at 85-95c for safe returns.

6. Cut Losers Fast If the thesis breaks, sell immediately. Recovering 40% beats losing 100%.


Output Format

For Crypto Signals:

BTC ^ UP  score=4 conf=60% | 5m momentum +0.12%, EMA5 > EMA10, last candle green
ETH - NEUTRAL score=1      | SKIP (weak signal)
SOL v DOWN score=-3 conf=45%| 5m momentum -0.08%, RSI 72 overbought, EMA bearish
XRP - NEUTRAL score=0      | SKIP (no signal)

ACTION: BET BTC UP $5.00 | BET SOL DOWN $3.00 | SKIP ETH, XRP

For Traditional Markets:

MarketClosesMarket PriceFair PriceEdgeConfidenceAction
.....................

Rules

  1. NEVER bet blind on crypto — TA score >= 3 or skip
  2. Always research traditional markets — never guess probabilities
  3. Be honest about uncertainty — if signals conflict, skip
  4. Show your work — display all indicator values
  5. Cut losers fast — sell if thesis breaks
  6. Never chase — missed entry? Wait for next window
  7. Log everything — track all trades for learning
  8. Patience = Profit — skipping weak signals IS a winning strategy

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.