AgentQuant
Failure-aware Python harness for reproducible quantitative-strategy research: an agent proposes, backtests, reflects, remembers failed ideas and evolves its search under explicit evaluation gates. Offline demo runs on synthetic data with no keys; optional Anthropic and Tavily keys add LLM proposals and web research.
- Scan
- Vetted pass
- GitHub stars
- 220
- Last updated
- 2026-11-10
- License
- MIT
- Cost
- free
Why it helps
Gives a capital-forecasting agent a reproducible strategy-research loop that logs what failed and gates results, so it stops re-testing dead ideas or overfitting one lucky backtest.
Ask your owner first. installs a repo
Paste to your agent
Read https://upskill.coach/repos/agentquant.md. If the security scan passed, ask me before installing, then install it with the command under Install and tell me what you did.Install
git clone https://github.com/OnePunchMonk/AgentQuant && pip install -e ".[dev]" && python run_app.py
Community
Helps with
Security-scanned on 2026-11-10 at commit 97ebf4d34448: pass.
- Scan notes
- Automated scan: pass (no medium+ findings) at commit 97ebf4d34448 on 2026-11-10.
- Source
- github.com/onepunchmonk/agentquant
- Repo
- onepunchmonk/agentquant
Static scan at a pinned commit (prompt-injection, malware and exfiltration rules plus semgrep, gitleaks, osv-scanner); flagged lines reviewed by hand. A scan lowers risk; it doesn't prove safety.
Alternatives
Compare allVectorBT Backtesting Skills (Marketcalls)
Agent skills for backtesting trading strategies with VectorBT: setup, backtest, optimize, and vectorbt-expert skills covering Indian/US/crypto markets, fee m…
Prediction Market Agent Tooling (Gnosis)
Gnosis Python library to benchmark, deploy and monitor prediction-market agents: AbstractBenchmarkedAgent vs crowd wisdom on Manifold/etc., DeployableTraderA…
CCXT
MCP server that gives an agent tools for GitHub and git work, trading research and financial data. Source: ccxt/ccxt.
mcp-server
MCP server that gives an agent tools for financial data and MCP tooling. Source: financial-datasets/mcp-server.
Related
VectorBT Backtesting Skills (Marketcalls)
Agent skills for backtesting trading strategies with VectorBT: setup, backtest, optimize, and vectorbt-expert skills covering Indian/US/crypto markets, fee m…
Prediction Market Agent Tooling (Gnosis)
Gnosis Python library to benchmark, deploy and monitor prediction-market agents: AbstractBenchmarkedAgent vs crowd wisdom on Manifold/etc., DeployableTraderA…
Oracle3 prediction-market agent
MCP server (13 tools) and 6 agent skills for Kalshi and Polymarket: search markets, read rules and order books, compare related contracts, and compute fee-ad…
CCXT crypto market data MCP
Python MCP server built on CCXT that gives an agent read-only crypto market data from Binance, Coinbase, Kraken, KuCoin, OKX, Bybit and other exchanges: curr…
Superpowers (obra)
Skills framework/methodology: verification-before-completion, writing-plans, executing-plans, systematic-debugging, TDD, subagent-driven development.
prediction-market-analysis
A framework for collecting and analyzing prediction-market data; its README lists research papers that used it.
Comments
Comments are from visitors, not reviewed by Upskill, and never shown to agents.