create forecast

Bind a forecast to a metric in your workspace. Stores the configuration, sets up the schedule, and resolves the source binding immediately so a bad source_id or

liveunknown~5s typical

mcpclaribi

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-com-claribi-mcp-server /create_forecast for me.

Input

Field	Type	Description
name*	string
description	string
source_type*	string (report \| dashboard \| goal \| data_source)	Which clariBI artifact the metric lives on. Most callers should start with data_source for
source_id*	string	UUID of the report, dashboard, goal, or data source.
metric_path*	string	Dot-path to the numeric column. For data_source: the metric key (e.g. "revenue"). For goal
granularity	string (hourly \| daily \| weekly \| monthly \| quarterly \| annual)	Bucket size for the time series. Hourly requires an hourly data source; the engine refuses default "daily"
horizon_days	integer	How many BUCKETS ahead to project. The per-granularity cap applies: hourly 168, daily 90, default 30
include_correlations	boolean	default true
include_anomalies	boolean	default true
include_changepoints	boolean	default true
narration_enabled	boolean	When true, every completed run gets an AI-generated narrative attached (summary, highlight default true
aggregation	string (auto \| sum \| mean \| last \| max)	How multiple raw points falling into the same bucket are combined. Auto picks sum for addi default "auto"
transform	string (auto \| none \| log)	Series transform applied before fitting. Log helps revenue/traffic series with growing var default "auto"
non_negative	boolean	When true, point + lower band clamped at zero. Omit to let the engine heuristic decide (vi
non_negative_auto	boolean	When true (default), the backend heuristic owns non_negative and re-evaluates per run base default true
method_override	string	Force a specific forecasting method instead of auto-selecting via walk-forward CV. Valid n
schedule_frequency	string (daily \| weekly \| monthly \| manual)	How often the forecast re-runs. Manual schedules only run when called explicitly via run_f default "monthly"
schedule_day_of_week	integer
schedule_day_of_month	integer

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-com-claribi-mcp-server","endpoint":"/create_forecast","input":{"name":"...","source_type":"report","source_id":"...","metric_path":"...","granularity":"daily","horizon_days":30,"include_correlations":true,"include_anomalies":true,"include_changepoints":true,"narration_enabled":true,"aggregation":"auto","transform":"auto","non_negative_auto":true,"schedule_frequency":"monthly"}}'

cli

npx cracked-ai run -p mcp-com-claribi-mcp-server -e /create_forecast -i '{"name":"...","source_type":"report","source_id":"...","metric_path":"...","granularity":"daily","horizon_days":30,"include_correlations":true,"include_anomalies":true,"include_changepoints":true,"narration_enabled":true,"aggregation":"auto","transform":"auto","non_negative_auto":true,"schedule_frequency":"monthly"}'

mcp

run_tool({ provider: "mcp-com-claribi-mcp-server", endpoint: "/create_forecast", input: {"name":"...","source_type":"report","source_id":"...","metric_path":"...","granularity":"daily","horizon_days":30,"include_correlations":true,"include_anomalies":true,"include_changepoints":true,"narration_enabled":true,"aggregation":"auto","transform":"auto","non_negative_auto":true,"schedule_frequency":"monthly"} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.