Forven // Instrument v0.1AGPL-3.0

Forge strategies. Trust only proof.

Forven is a free, open-source, self-hosted workspace where AI research agents, a backtesting gauntlet, and a risk command center turn fragile ideas into evidence — local-first, your keys, no account.

Clone the repo →See the method ↓
  • Local-first
  • BYO keys
  • No account
  • AGPL-3.0
Lab snapshot — daemon healthySample
Gauntlet attrition — survivors per stage1,000 → 2
screen280/1,00028%
gauntlet42/28015%
candidate9/4221%
live2/922%
App // DashboardSample
Forven operator dashboard — ALL SYSTEMS GO: system pulse, scheduler queue, and an open-trades table across ETH, BTC, AVAX, and APT.
Mode: paper
Local-first

Strategies, data, and runs stay on your machine.

Bring your own keys

Your exchange and LLM keys never leave your machine.

No account

Clone and run — no sign-up, no license key.

AGPL-3.0

Free and open source. Read every line.

01What you doLocal process

A research process with hard gates.

Forven is not a signal feed. It is a local process for turning fragile ideas into candidates that have earned further attention.

01LAB

Build the cohort

Collect strategy ideas, parameter sets, and agent-drafted variants in one local lab before anything graduates.

02GAUNTLET

Run the gauntlet

Push each idea through screen, walk-forward gauntlet, candidate, and live-review stages with explicit gates.

03PROMOTE

Promote with evidence

Use out-of-sample metrics, audit trails, and paper-trade behavior to decide what survives and what gets retired.

02The gauntletWalk-forward · Sample

Most strategies fail. Forven makes failure useful.

Each stage asks a harsher question of the same idea. If the out-of-sample evidence stops holding up, the strategy stops moving forward.

Attrition — survivors per stage (of 1,000)Sample
screen280/1,00028%

Fast filters remove fragile ideas before deeper testing.

gauntlet42/28015%

Walk-forward windows test survival outside the fit period.

candidate9/4221%

Paper trading and review catch behavior that backtests hide.

live2/922%

Only a small set earns live-routing consideration.

0.2%
Graduation ratio
Illustrative cohort math, not a promise about your universe, regime, or results.

The point is not optimism.

The point is disciplined rejection. Forven helps you learn why an edge died before that lesson costs live capital.

App // The CrucibleSample
Forven Crucible — the research inbox: hypotheses with novelty scores, scope, and out-of-sample results, tagged TESTED, AGENT, and OPERATOR.
Mode: paper
03Risk commandSelf-halting · Sample

A risk command center that can halt itself.

Drawdown, daily-loss, and portfolio-risk limits run as live gauges with circuit breakers. Cross a threshold and Forven blocks new entries — before you do.

Risk limits — current vs. ceilingSample
1.98%
Drawdown
limit 30%
0.94%
Daily Loss
limit 5%
1.00%
Portfolio Risk
limit 2%
Circuit breakers
Price Feed: closedTrade: closedAccount: closed

Breakers gate the price feed, trade routing, and account before a bad fill compounds.

One control freezes all execution and parks open work for review.

App // Risk CommandSample
Forven Risk Command — drawdown 1.98%, daily-loss 0.94%, and portfolio-risk 1.00% gauges with circuit breakers closed and an emergency-halt control.
Mode: paper
04Inspect the labDatasheet · Sample

Inspect the lab before you trust the output.

Forven keeps the workflow visible: strategy cohorts, current stages, recent out-of-sample behavior, and the trail of decisions that moved an idea forward.

Strategy cohortSample
IDStrategyStageTradesOOSStatus
S00241rsi_mean_reversion_btclive127+2.4%pass
S00512vwap_reclaim_ethcandidate38+0.9%pass
S00693volume_anomaly_solgauntlet212-1.1%fail
S00718breakout_confirm_btcgauntlet94+0.4%watch
Trust surface
  • Local by default
    Strategies, data, and runs stay on your machine unless you choose to connect external services.
  • Bring your own keys
    Use your own exchange and LLM keys. Forven does not need to resell model tokens to be useful.
  • Auditable gates
    Every promotion has a stage, a metric trail, and a reason. No hidden signal feed, no black-box handwave.
App // The ForgeSample
Forven The Forge — the strategy cohort leaderboard with OOS CAGR, OOS Sharpe, Max DD, and stage columns across Open, Parked, and Graveyard tabs.
Mode: paper
05Paper tradingTestnet · Sample

Paper-trade on testnet before any live capital.

Survivors run forward on Hyperliquid testnet with the same risk controls a live account would use — so behavior that backtests hide shows up where it’s safe.

Hyperliquid testnet

Paper-trade against a real venue, no real capital at risk.

Real risk controls

Stop-losses, a drawdown kill-switch, and liquidation-distance monitoring.

Fill reconciliation

Every simulated fill is reconciled so the metrics can't quietly drift.

App // Paper TradesSample
Forven Paper Trades — a live session for an ETH/USDT strategy with candlestick price and indicator charts, signals, and an open-position panel.
Mode: paper

Sample session for illustration — strategy names, charts, and PnL are not real or projected returns.

06AgentsFree & open source

Co-researchers, not co-pilots.

Agents are optional co-researchers included for everyone. They can draft hypotheses, inspect failed runs, and explain drawdowns, but the gauntlet still decides what earns attention.

You bring your own LLM key for model calls. The lab, daemon, and agent layer are all free and open source.

  • Drafts hypotheses
    from natural-language research prompts
  • Edits strategy code
    against your actual repo, with diffs
  • Inspects gauntlet results
    reasons about why edges died
  • Writes post-mortems
    on live drawdowns and failed candidates
How the agent layer works →
App // Agent HubSample
Forven Agent Hub — a grid of idle agent cards (Quant Researcher, Simulation Agent, Risk Manager, Execution Trader, and more), each configured with a per-agent model.
Mode: paper
07Get ForvenAGPL-3.0

Free and open source.

Every Forven feature is free and open source. Clone the repository and run one launcher script — the whole stack runs on your own machine. Bring your own LLM key when you want agents to help draft, debug, and review.

Clone the repo →Read the install guide →Join the Discord →
  • Clone the repo and run one launcher script — no installer, no account.
  • Bring your own exchange and LLM keys; nothing is resold or proxied.
  • Runs on Windows, macOS, and Linux from source.