Causal Decision Agent

Overview Schema Source Related Servers Discussions Admin

arima

Read-only

Fit seasonal time-series models with exogenous regressors to forecast outcomes and support causal decisions. Returns validated parameter estimates and diagnostics with audit-ready evidence.

Instructions

Fit ARIMA(p,d,q) or SARIMAX. Validation: certified parity evidence.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`y`	Yes	Outcome variable column name or outcome array.
`auto`	No	If True, select (p, d, q) by AICc grid search (ignores ``order``).
`exog`	No	Exogenous regressors (ARIMAX).
`max_d`	No	Bounds for the auto search.
`max_p`	No	Bounds for the auto search.
`max_q`	No	Bounds for the auto search.
`order`	No	order parameter (Tuple[int, int, int]).
`detail`	No	Payload depth: 'minimal' (~150 tokens) for sub-step calls where only the point estimate is needed; 'standard' (~1K tokens) for diagnostics + coefficient table; 'agent' (~2K tokens, default) adds violations / next_steps / suggested_functions so the LLM can plan its next call without another round-trip.	agent
`method`	No	Estimation convention. ``'statespace'`` keeps the default exact Kalman/SARIMAX likelihood. ``'css_ml'`` is retained as a compatibility alias for ``'innovations_mle'``. The innovations-MLE path uses statsmodels' stationary/invertible exact-MLE parameterization, matching ``stats::arima(method='ML')`` and tightly converged Stata ``arima`` coefficient conventions for pure ARMA models.	statespace
`as_handle`	No	If true, cache the fitted result on the server and return result_id + result_uri alongside the JSON payload so a subsequent tools/call can chain without re-running.
`data_path`	No	Absolute path or URL to a data file. Supported: .csv / .tsv / .txt (delimited), .parquet / .pq, .feather / .arrow, .xlsx / .xls, .dta (Stata), .json / .jsonl. Schemes: file://, s3://, gs://, https://.
`result_id`	No	Optional handle to a previously-fitted result (returned by an earlier call when as_handle=true). Tools that operate on a fitted object accept this in place of re-supplying data_path + columns.
`data_columns`	No	Optional column projection. Parquet/Feather/Stata loaders honour this for fast partial reads.
`data_sample_n`	No	Optional uniform random subsample size (seed=0, deterministic) — useful on huge panels.
`seasonal_order`	No	seasonal_order parameter (Optional[Tuple[int, int, int, int]]).

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 15, 2026

TDQS

B3.1/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations provide readOnlyHint=true and openWorldHint=false, so the description is not burdened with declaring that fitting is a read-only computation. The description adds only the vague phrase 'certified parity evidence,' which does not clarify what is validated or against which reference. No contradiction with annotations exists.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is very short and front-loads the core purpose. However, the second sentence, 'Validation: certified parity evidence,' is cryptic and does not earn its place because it is not actionable or clearly explained.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 15-parameter fitting tool with auto-search, exog, seasonal_order, method variants, and handle-based caching, the description is thin. The rich schema and output schema compensate for invocation details, but the description alone gives an agent little context about when this tool is the right choice or what 'certified parity evidence' means.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the parameter meanings are already fully documented in the schema. The description adds no parameter semantics beyond hinting at SARIMAX, which maps to the seasonal_order parameter already described. A baseline score of 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific action and resource: 'Fit ARIMA(p,d,q) or SARIMAX.' This is clear and distinguishes the tool from the many non-time-series siblings. However, it does not explicitly contrast with close model-fit siblings like garch or var, so it stops short of full differentiation.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

There is no guidance on when to use this tool versus alternatives, nor any indication of prerequisites or preferred workflows. The cryptic 'Validation: certified parity evidence' does not help an agent decide when to call arima instead of garch, var, or regress.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server