Install
openclaw skills install silk-manifoldSearch, analyze, and trade on Manifold Markets prediction markets by checking odds, placing bets, viewing portfolios, and managing positions with API-based c...
openclaw skills install silk-manifoldTrade on prediction markets via the Manifold API. All commands output JSON.
Script path: {baseDir}/scripts/manifold.py
Run with: python3 {baseDir}/scripts/manifold.py <command> [options]
python3 {baseDir}/scripts/manifold.py balance
# Search by keyword
python3 {baseDir}/scripts/manifold.py search "US election"
# Open binary markets, sorted by popularity
python3 {baseDir}/scripts/manifold.py search "AI" --filter open --sort most-popular --limit 5
# Closing soon
python3 {baseDir}/scripts/manifold.py search "" --filter closing-week --sort close-date
Filters: all, open, closed, resolved, closing-day, closing-week, closing-month
Sorts: most-popular, newest, score, daily-score, 24-hour-vol, liquidity, close-date, prob-descending, prob-ascending
# By ID
python3 {baseDir}/scripts/manifold.py market <market-id>
# By slug (from URL)
python3 {baseDir}/scripts/manifold.py market some-market-slug
# Just the probability
python3 {baseDir}/scripts/manifold.py prob <market-id>
# Market order: bet 100 mana on YES
python3 {baseDir}/scripts/manifold.py bet <contract-id> 100 YES
# Limit order at 40% probability
python3 {baseDir}/scripts/manifold.py bet <contract-id> 100 YES --limit-prob 0.40
# Dry run (simulate without executing)
python3 {baseDir}/scripts/manifold.py bet <contract-id> 100 YES --dry-run
# Bet on a specific answer in a multiple-choice market
python3 {baseDir}/scripts/manifold.py bet <contract-id> 50 YES --answer-id <answer-id>
# Limit order that expires in 1 hour
python3 {baseDir}/scripts/manifold.py bet <contract-id> 100 YES --limit-prob 0.35 --expires-ms 3600000
# Sell all shares in a market
python3 {baseDir}/scripts/manifold.py sell <contract-id>
# Sell specific outcome
python3 {baseDir}/scripts/manifold.py sell <contract-id> --outcome YES
# Sell partial shares
python3 {baseDir}/scripts/manifold.py sell <contract-id> --outcome NO --shares 50
# Sell in multiple-choice market
python3 {baseDir}/scripts/manifold.py sell <contract-id> --answer-id <answer-id>
python3 {baseDir}/scripts/manifold.py cancel <bet-id>
# Your portfolio summary
python3 {baseDir}/scripts/manifold.py portfolio
# Your current positions with contract details
python3 {baseDir}/scripts/manifold.py my-positions --limit 10 --order profit
# Positions/leaderboard for a specific market
python3 {baseDir}/scripts/manifold.py positions <contract-id> --top 10
# Your position in a specific market
python3 {baseDir}/scripts/manifold.py positions <contract-id> --user-id <your-user-id>
# Your recent bets
python3 {baseDir}/scripts/manifold.py bets --username <your-username>
# Bets on a specific market
python3 {baseDir}/scripts/manifold.py bets --contract-id <contract-id>
# Open limit orders only
python3 {baseDir}/scripts/manifold.py bets --username <your-username> --kinds open-limit
python3 {baseDir}/scripts/manifold.py me
--dry-run first when the user asks to place a bet, so they can confirm the expected outcome before committing real mana.search results include id fields — use these as contract-id for betting/selling.answers and their IDs.