Tools and Resources

Everything the MCP server exposes: tools by area, the resources you can pull by name, and the response envelope every tool shares.

The response envelope

Every tool answers the same shape:

{ "ok": true, "tool": "get_metric_data", "data": { }, "error": null, "errorKind": null, "details": null }

Read the named collection inside data — data.metrics, data.items, and so on. When ok is false, errorKind says which kind of refusal it is (see when it refuses).

No ambient context. Unlike the web chat, there is no open chart and no conversation history: pass assetId and chartId explicitly on every call. Resolve them first with search_assets, suggest_charts / search_charts and search_metrics.

Discovery and metrics

Tool	What it does
search_metrics	Find a metric by name or concept
get_metric_info	Details of a single metric
get_metric_data	Time-series for a metric and asset
compare_metrics	Two or more metrics on the same window
get_metrics_variation	What moved today
get_endpoint_parameters	Which parameters a metric accepts, including its timeframes

Assets and market

Tool	What it does
search_assets	Resolve an asset by name or ticker
get_asset_info	Asset details
get_summary_info	Market summary for an asset
get_asset_price_history	Price history
get_trending_assets	Trending assets over a period (1d, 7d, 30d)

Charts

Tool	What it does
suggest_charts	Charts relevant to a question
search_charts	Find a chart by name
get_chart_details	What a chart is and how it is built
get_chart_metrics	Metrics plotted on a chart
compare_charts	Two charts side by side
get_charts_by_asset	Charts covering an asset
get_charts_by_category	Charts in a category
get_categories_by_asset	Categories available for an asset

Analysis

Tool	What it does
backtest_metric	What happened historically after a metric crossed a threshold
rank_assets_by_metric	Rank assets by any metric
compute_purchase_plan	Build a staged purchase plan

Screener, derivatives and signals

Tool	What it does
get_screener_data	Screener rows
get_screener_overview	Screener ranking
get_liquidation_clusters	Liquidation clusters for an asset and timeframe
get_ohlcv_spot	Spot candles
get_signal_series	Trading indicator series — the indicator is an argument
get_signal_availability	Which buckets an asset actually has

Research

Tool	What it does
search_reports	Search research by content
get_reports_by_asset	Reports covering an asset
get_latest_reports	Most recent reports
get_report_key_insights	Key insights from a report

Reports above your plan come back as metadata only, flagged with accessDenied and requiredPlanTier. That is a plan wall, not an empty report.

Your account

Tool	What it does
get_my_plan	Your plan and what it entitles you to
get_my_usage	Credits and rate limit consumed
get_user_favorites	Your favorite assets
get_user_alerts	Your alerts
get_alert_triggers	When your alerts fired
check_alert_usage	How many alerts your plan allows and how many are used

Tools that change something

These are flagged in the protocol so a well-behaved client asks you first.

Tool	What it does
prepare_alert	Stage an alert and return the exact configuration plus a token
confirm_alert_creation	Create the alert that was staged, by token
update_alert	Edit an existing alert
delete_alert	Delete an alert
add_favorite_asset / remove_favorite_asset	Manage favorites
list_screener_views	Your saved screener views
create_screener_view / update_screener_view / delete_screener_view	Manage saved views

Alert creation is deliberately two calls. prepare_alert does not create anything — it returns the configuration for you to read and a token. Only confirm_alert_creation with that token creates the alert, and what gets created is what you were shown.

Resources

Pullable by name, without a tool call:

Resource	URI
Metric catalog	alphractal://catalog/metrics
Asset catalog	alphractal://catalog/assets
My favorites	alphractal://user/favorites
Latest research reports	alphractal://reports/latest
A specific report	alphractal://reports/{id}

Prompt

asset-deep-dive — a structured prompt that runs the tools in an order that works, instead of leaving the model to guess the sequence.

Timeframe is a resolution, never a period

One vocabulary across the whole platform — 1m, 5m, 15m, 30m, 1h, 4h, 6h, 12h, 1d — in which 1m is one minute. There is no weekly or monthly interval anywhere.

timeframe sets the spacing between points; the period is startDate and endDate. Where a tool ranks or measures over a span instead of sampling a series, the argument is called period and takes spans (1d, 7d, 30d) — get_trending_assets is the one that does. So 7d is always a span and never a spacing.