v0.4.3 — home in the varrdinc org

VARRD's public repo now lives at github.com/varrdinc/varrd (moved from the personal namespace; old links redirect).

• All package and listing URLs point at the org repo
• MCP registry identity is now io.github.varrdinc/varrd
• Refreshed PyPI / Smithery / Glama listing copy; contact is management@varrd.com
• Clarified free-credit terms in the balance skill

No functional changes — CLI, SDK, and MCP behavior identical to 0.4.2.

v0.4.2

v0.4.2

v0.4.1

v0.4.1: Clearer onboarding — Manage Agents flow, merge agent, email v…

v0.4.0 — Edge Library + Full Performance Analytics

What's new

Edge Library (varrd_edges)

• Browse validated trading edges firing in real time
• Three tiers: free (markets), $0.50 (stats + trade levels), $1/edge (full audit trail)
• Filters: direction, timeframe, asset_class, market, status
• Section drill-downs: horizons, analytics, occurrences, setup_code, view

Full Performance Analytics

• SQN, profit factor, Kelly %, payoff ratio, streaks
• Monte Carlo simulation (500 runs, profitability %, percentiles)
• Drawdown analysis (max DD, recovery factor)
• Return distribution (skew, kurtosis, tail ratio)
• Regime analysis (VIX + market trend breakdowns)
• Edge decay by quarter, rolling stability score

SDK + CLI

• New edges() method with direction, timeframe, asset_class filters
• New varrd edges CLI command
• MCP stdio proxy: fixed auth + timeout

Security

• Webhook signature verification
• Atomic payment idempotency
• Cross-user payment replay blocked
• Batch + body size limits on MCP endpoint

Cleanup

• Full README rewrite with screenshots
• Removed stale files (AGENTS.md, scan skill)
• Updated all registry configs

v0.3.4

v0.3.4 — Add MCP stdio proxy + mcp-remote for LobeHub validation

v.0.3.3

v0.3.3: Add glama.json, bump version, gitignore token files

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

v0.3.2 — Get Briefed

Add varrd briefing command — personalized market news briefing based on your edge library

v0.3.1 - Jesse integration

Jesse trading bot integration

varrd.jesse generates ready-to-run Jesse Strategy files from VARRD-validated hypotheses -- same pattern as varrd.freqtrade.

New

• generate_strategy() -- Jesse Strategy class + routes.py with ATR SL/TP, risk-based sizing, time exits
• validate_strategy() -- parse existing Jesse strategies and validate through VARRD
• Translator -- markets (BTC_daily -> BTC-USDT), timeframes (daily -> 1D), indicators, entry logic

Usage

from varrd.jesse import generate_strategy
strategy_code, routes_code = generate_strategy(hypothesis, setup_code=code)

pip install --upgrade varrd

v0.3.0 — Freqtrade integration

Freqtrade Integration

VARRD now generates ready-to-run Freqtrade strategy files from validated edges. No more coding strategies blind — validate the idea
first, then deploy it.

What's new

from varrd import VARRD
from varrd.freqtrade import generate_strategy

v = VARRD()
result = v.discover("RSI oversold reversal on BTC")

if result.has_edge:
hyp = v.get_hypothesis(result.hypothesis_id)
strategy_code, config = generate_strategy(hyp)

  with open("VARRDStrategy.py", "w") as f:
      f.write(strategy_code)
  # Drop into Freqtrade and run it

What it does

• generate_strategy() — Takes a VARRD-validated hypothesis and generates a complete Freqtrade .py strategy file with proper
  populate_indicators(), populate_entry_trend(), ATR-based stop loss, take profit, and time-based exits. Includes full validation stats
  in the header.
• validate_strategy() — Already have a Freqtrade strategy? Parse it and validate through VARRD to check if the edge is real.

v0.2.0

What's New

• buy_credits tool — CLI (varrd buy-credits) + SDK (v.buy_credits())
• Welcome banner with passkey, balance, and merge instructions for new agents
• SDK handles multi-content responses (banner + tool data)
• Updated AGENTS.md with credits, payment, and account linking docs
• Updated instructions with buy-credits command and account linking flow

Install / Upgrade

pip install --upgrade varrd