app.nothingbutalpha/backtest-auditor

What it mainly works on, as our AI model read its tool descriptions: files and code on the machine it runs on; nothing outside itself: it only answers questions.

None of its 6 tools are described as changing anything.

• check_codeon 6 other serversruns code or commands

  “Static review of Python/pandas strategy code for patterns that inflate backtests: negative shifts in features or signals, centered windows, backward fill, fitting before the time split, shuffled CV, …”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  1 input, 1 required

  • sourcestringrequired“Python source of the strategy/backtest (pandas, numpy, sklearn, vectorbt, backtrader...).”
• check_pathon 1 other serverreads or searches

  “Read-only lint of Python strategy files on disk: the same rules as check_code, per file.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  3 inputs, 1 required

  • pathstringrequired“A .py file, a .ipynb notebook, or a folder to scan.”
  • include_globstring or null“Optional filter on paths relative to the folder, e.g.”
  • max_filesinteger or null“Most files to read (default 200, at most 2000).”
• audit_backteston 2 other servers

  “One call that runs every audit that applies: check_path on the code, audit_returns on the returns, audit_trade_log on the trades, all against the benchmark if given.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  8 inputs

  • code_pathstring or null“Strategy code: a .py file, notebook or folder (absolute path).”
  • returns_csvstring or null“Path to the strategy's per-period returns (CSV date,return or JSON).”
  • trades_csvstring or null“Path to a trade log (date, ticker, side, qty, price[, fee]).”
  • benchmark_csvstring or null“Path to benchmark returns, same format as returns_csv.”
  • n_trialsinteger or null“Total strategy variants tried before choosing this one.”
  • turnovernumber or null“Average one-way turnover per period, for the cost check.”
  • costs_bpsnumber or null“All-in cost per unit of turnover, in bps.”
  • prices_csvstring or null“Optional daily closes for marking the trade log.”
• audit_returns

  “Statistical red-team of a backtest's return series: net vs gross at 0/5/10/20 bps and the break-even cost; Sharpe with Newey-West and Lo-style t-stats (autocorrelation-aware); the Deflated Sharpe Rat…”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  11 inputs, 1 required

  • returnsstringrequired“Per-period strategy returns as CSV (date,return) or JSON (list of numbers, {date: return}, or records).”
  • freqstring“Return frequency: auto (inferred from the dates; default), D, W, M, Q, Y, or periods per year as a number.”
  • costs_bpsnumber or null“Your all-in cost per unit of one-way turnover, in bps.”
  • turnoverstring or number or null“One-way turnover per period (sum of |change in weights|): a number (average) or a series like `returns`.”
  • positionsstring or null“Optional weights table (CSV/JSON): wide date + one column per asset, or long date,ticker,weight.”
  • benchmarkstring or null“Optional benchmark returns, same format as `returns`.”
  • n_trialsinteger or null“Total number of strategy variants tried before choosing this one (parameters, universes, rules).”
  • trials_sr_variancenumber or null“Optional variance of the annualized Sharpe ratios across all trials.”
  • input_kindstring“'returns' or 'equity' (an equity/NAV curve).”
  • returns_are_netboolean“True if the returns already include trading costs.”
  • unitsstring“'decimal' (0.01 = 1%) or 'percent' (1 = 1%).”
• audit_trade_log

  “Builds daily PnL, returns and turnover from a trade log (marked to market with your prices if given), then runs the full audit_returns review.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  6 inputs, 1 required

  • tradesstringrequired“Trade log as CSV or JSON with columns date, ticker, side (buy/sell/short/cover), qty, price, and optionally fee.”
  • pricesstring or null“Optional daily closes for marking positions: wide (date + one column per ticker) or long (date,ticker,close).”
  • initial_capitalnumber or null“Starting capital.”
  • costs_bpsnumber or null“Cost per unit of turnover in bps for the headline result (if the log has no fee column).”
  • benchmarkstring or null“Optional daily benchmark returns (CSV/JSON).”
  • n_trialsinteger or null“Total strategy variants tried before this one.”
• explainon 29 other serversreads or searches

  “What a check looks for, why the problem inflates backtest results, how to fix it, and references.”

  Publisher declares: read-only yes · destructive no · idempotent yes · open-world no

  1 input

  • check_idstring“Check id, e.g.”

Each quoted sentence is the first sentence of the publisher’s own description. The kind of tool and the “changes data” mark are our AI model’s reading of that description, not something we tested: two differently worded runs had to agree, otherwise no label is shown, and nobody has checked them by hand. A tool without the “changes data” mark is described as only reading, searching or calculating, or the two runs did not agree. No tool was called. Read the publisher’s full description before relying on it. “Publisher declares” lists the hints the server itself sends with the tool (MCP annotations: read-only, destructive, idempotent, open-world); they are the publisher's claims, not checked by us. Inputs are the parameters in the tool's own input schema: name, declared type, whether it is required, and the first sentence of its description. Read 2026-10-23.

The registry verifies a package once, when it is published, and does not re-check it. A difference below is about the listing, not a judgement of the publisher.