Research

The most comprehensive MCP server for Polymarket — 48 tools spanning direct trading, market discovery, smart money tracking, copy trading, backtesting, risk management, and portfolio optimization. Wor

AI prediction market trading agent scanning Polymarket for alpha using microstructure research, regime detection, and sentiment analysis.

Funding intelligence workflow using Messari x402. Scans recent rounds, M&A activity, venture funds, and organizations, then synthesizes a VC landscape brief via Messari AI. Total cost ~$0.75–$1.25 USDC per run.

Use when the user asks to "fix false positives", "check FP stats", "triage autofixer false positives", "why is rule X getting dismissed", or wants to reduce dismissal rates in the program_autofixer. Queries live dismissal telemetry from Databricks, ranks rules by false-positive burden, diagnoses the root cause in the rule's visitor, and implements a fix with the user approving each step.

Paper-trade on CoinRithm via the coinrithm-trading MCP server. Use when the user wants to paper trade, check their CoinRithm portfolio, get a price/quote, open or close a position (spot, futures, or prediction markets), set a stop-loss/take-profit, place or cancel an order, check balances/PnL, or see the public Agent Arena leaderboard on CoinRithm. All trading is simulated virtual funds — never real money.

Validate review findings about an implementation plan against the codebase and repo rules, fix the valid ones in the plan, and write a feedback file noting what changed and what was rejected (and why). For plan-review findings, not code-review findings.

Search and discover 43k+ AI agents registered via ERC-8004. Find agents by skill, chain, or reputation. View leaderboards, ecosystem stats, and monitor metadata changes.

Use this skill for meme/打狗/alpha token research on pump.fun and similar launchpads: scanning new token launches, checking developer reputation/开发者信息/dev launch history/has this dev rugged before/开发者跑路记录, bundle/sniper detection/捆绑狙击, bonding curve status/bonding curve progress, finding similar tokens by the same dev/相似代币, and wallets that co-invested (同车/aped) into a token. Use when the user asks about 'new meme coins', 'pump.fun launches', 'trenches', 'trench', '扫链', 'developer launch histor...

Use this skill for token-level data: cluster overview, search tokens, trending/hot tokens (热门, 代币榜单), liquidity pools, holder distribution (whale/巨鲸, sniper, bundler percent), token safety and honeypot risk (貔貅盘, 'is this token a honeypot', 'is this safe to buy', 'rugged?', 'can I sell this'), who created a token, recent buy/sell activity, top profit addresses, trade history by wallet type, or holder cluster analysis (持仓集中度, rug pull probability/跑路风险, new wallet percentage/新钱包持仓比例, suspicious...

Use this skill for token-level data: search tokens, trending/hot tokens (热门, 代币榜单), liquidity pools, holder distribution (whale/巨鲸, sniper, bundler-tagged holder %), token risk metadata (riskControlLevel, tokenTags, dev stats, suspicious/bundle holding % via advanced-info), recent buy/sell activity, trade feed/逐笔成交/每笔交易/stream trades, top profit addresses, token trade history, detailed price info with market cap volume liquidity and holder count (price-info), or holder cluster analysis (持仓集中度...

Analyze a crypto trade before or after execution to identify risk, leverage, sizing, liquidity, volatility, timing, and execution problems using Binance market data.

MCP server for Lighter Protocol — graded safety gates (readonly/paper/live/funds), audit log, two-step confirmations.

Example driver: drive a portfolio-rebalance hypothesis into a recorded session.

Paper-trading sandbox over workspace sessions: record live market data and agent decisions inside a session window, replay and score it later.

Run the Autoresearch Lab lifecycle end-to-end: freeze a hypothesis with mechanical success criteria and a sealed session plan, drive recorded and replayed paper runs, score them honestly, and read the verdict — resumable from disk at every step, designed for /loop.

Automated smart contract security auditor with ZK-proven assertions (Groth16) and multi-scanner AI synthesis.

Use when drafting, restructuring, or polishing a cryptography paper for a specific top venue (CRYPTO, EUROCRYPT, ASIACRYPT, IEEE S&P, ACM CCS, USENIX Security, NDSS) or for CRYPTO/LNCS-style submission, including venue structure, abstract compression, de-AI rewriting, double-blind compliance, and format checks.

Use when auditing a cryptographic theorem, lemma, proof, game hop, reduction, ZKP security property, lattice assumption, or claimed protocol-security implication.

Use when orchestrating an end-to-end cryptography, zero-knowledge proof, or lattice research project from an idea through literature, novelty, proofs, parameters, experiments, writing, internal review, and submission readiness.

Use when synthesizing a verified cryptography, ZKP, or lattice paper corpus into a taxonomy, historical development, claim comparison, tradeoff map, research gaps, or related-work section evidence base.

Use when outlining, drafting, restructuring, or completing a cryptography, ZKP, lattice, or security paper from verified claims, proofs, literature, parameters, experiments, artifacts, and venue requirements.

Use when pre-reviewing, peer reviewing, stress-testing, or assessing submission readiness of a cryptography, ZKP, lattice, or security paper for a leading research venue.

Use when deeply reading one cryptography, ZKP, or lattice paper to extract its claims, definitions, proof structure, assumptions, parameters, evidence, costs, limitations, and research implications.

Use when conducting cryptography, ZKP, or lattice literature discovery, prior-art mapping, citation verification, closest-work comparison, related-work evidence collection, or bounded novelty searches.