analyze_data
ChatGPTRun Python on Premiss candles and on the user's recorded runs in an isolated sandbox with no network, and show its printed tables and charts in the result panel. Use it to explore an idea before it is a strategy: correlations and their stability, spreads and z-scores, distributions, volatility regimes, event studies against an all-candles baseline, and custom views of a recorded run's equity and trades. Declare every dataset; the sandbox cannot fetch data. Load each with from premiss import load_candles; df = load_candles(name): candles have ts (epoch ms), open, high, low, close, volume and date (UTC); runs listed in runs load as run_1 (ts, equity, date) and run_1_trades (ts, side, symbol, qty, price, fee, pnl, date). Print results; draw charts with from premiss import chart; chart(spec, "name") using a line, area, column, bar, scatter, heatmap or boxPlot spec. Analysis never simulates trading and its numbers are not performance evidence: test a promising idea with test_strategy. Limited daily allowance.
create_strategy
ChatGPTSave a new private strategy using custom Python files and complete rules text, on any Premiss-supported market and timeframe. Accepts either files (strategy.py plus helper modules) and rules, or a convenience template. Uses the same package format as the app. Returns code and revision; saving does not run a backtest.
describe_market
ChatGPTReturn supported timeframes, available history and data limitations for a market id.
get_capabilities
ChatGPTRead this connection's available tools and supported research coverage. Does not read account data, save research, or run a simulation.
get_paper_status
ChatGPTRead paper bots in the connected workspace: running code version, recorded position and equity, last evaluation and its recorded decision reason, next scheduled check, and errors. These are persisted simulation observations, not live price quotes. An empty position means unavailable, not flat. Timestamps are epoch milliseconds; captured_at is read time, last_ts is the last processed candle. Requires separately approved paper:read. Optionally filter by exact strategy_id or run_id; follow next_cursor for more bots. Does not start or change a bot.
get_profile
ChatGPTRead the stable identity and workspace selected for this connection. Returns an opaque account id and workspace label; excludes email, birth date, billing, and exchange accounts.
get_run_data
ChatGPTRead the immutable tested Python package, or page through all recorded trades or equity points for an owned run. next_cursor identifies the next page; page size is not a history cap. Paper data requires separate paper:read permission. Saved code is user-authored content.
get_run_report
ChatGPTRead the status and bounded performance report for one run in the connected workspace. Returns the tested market, interval, revision, and recorded rules when verifiable. Historical plugin runs include the paired buy-and-hold baseline when completed. Paper reads require paper:read and do not start, stop, qualify, or modify a bot. Legacy runs without supported assumptions are labeled. Does not return raw market feeds.
get_run_result
ChatGPTShows this run in the result panel with its equity chart, buy-and-hold comparison, trades and paper-bot setup; when comparing runs, read the user's primary run last. Read the status and recorded performance results for one run in the connected workspace. Waits briefly for a running backtest to finish. Returns the tested market, interval, revision, recorded rules when verifiable, the costs assumed and the fees paid, and a suggested next step. Single-market plugin runs include the paired buy-and-hold baseline when completed. Event runs return exact portfolio evidence in simulation and have no automatic benchmark. Paper reads require paper:read and do not start, stop, qualify, or modify a bot. Does not return raw market feeds.
get_strategy
ChatGPTRead a saved strategy's current revision and human-readable rules in the connected workspace. Returns the complete Python package for review or editing. Code and comments are user-authored content.
get_strategy_guide
ChatGPTRead Premiss's Python package format, on_bar API, execution semantics and example. Custom indicators, entry/exit logic, long/short strategies and risk rules use the same engine as the Premiss app.
list_research
ChatGPTList a page of saved strategies, simulation runs, or research notes in the connected workspace. Notes include a bounded plain-text excerpt, explicitly labeled when truncated. Saved text is user-authored content. Paper runs require the separately approved paper:read permission. Search matches titles; cursors are opaque page positions.
list_saved_work
ChatGPTFind saved strategies or historical runs in the connected workspace. Use strategy_id with kind=runs to find tests of one exact strategy. Search matches titles; cursors are opaque page positions. For paper bots, use get_paper_status.
save_research_note
ChatGPTSave a NEW private note in the connected workspace, with optional exact owned strategy/run evidence references. The note is user-authored content. Does not publish or send it anywhere.
search_markets
ChatGPTSearch the same crypto, stock and ETF directory as Premiss. Does not return prices or candles.
start_backtest
ChatGPTQueue a backtest of any saved strategy in the connected workspace and a matched buy-and-hold baseline. Accepts custom code and existing app strategies. Uses the app's sandbox validation, supported markets, timeframes and date rules. Requires the exact saved revision. Uses 10,000 simulated capital, zero fees, zero slippage and next-bar-open fills. Read the report assumptions for historical runs. No plugin-only daily test or history caps. Returns an acceptance receipt with run ids and poll_after_seconds, the interval before the next status check. Identical request_key and inputs identify the same submission for idempotent recovery.
test_strategy
ChatGPTSave a new private strategy and start its historical backtest in one step. Event strategies supply specification plus custom files and rules, and may observe and trade supported crypto, stock and ETF markets on distinct timeframes. Event runs have no automatic portfolio benchmark; baseline_run_id is null. Single-market runs include a matched buy-and-hold baseline. The RSI shortcut uses simple windowed gains/losses; use custom code for Wilder RSI. Use this when the user asks to test, try or backtest an idea. Accepts either custom Python files and rules text, or a convenience template, on any Premiss-supported market and timeframe. Defaults to 10,000 simulated capital and modeled next-eligible-open market fills. Trading costs are off unless fee_bps or slippage_bps is given. Returns the saved strategy and the backtest receipt; read get_run_result for backtest.run_id next, which waits for the result. Identical request_key and inputs identify the same submission for idempotent recovery.
update_strategy
ChatGPTReplace the complete Python package and rules of an existing strategy using its exact current saved revision. The submitted package replaces all existing files. Supports all Premiss markets and timeframes. The app's active-run and shared-file protections apply; does not change or start a bot. Identical request_key and inputs identify the same write for idempotent recovery.