Causal Decision Agent

Overview Schema Source Related Servers Discussions Admin

sensitivity_rr

Read-only

Assess causal estimates' robustness to parallel-trends violations via Rambachan-Roth honest-DiD sensitivity analysis. It computes breakdown Mbar (largest allowed violation) and robust confidence sets for the post-treatment ATT.

Instructions

Rambachan-Roth (2023) honest-DiD sensitivity analysis: computes the largest violation of parallel trends (parametrised by Mbar -- relative magnitude of the post-period violation versus the worst observed pre-period one) under which the post-treatment ATT is still different from zero at level alpha. Reports both the robust confidence sets and the breakdown Mbar. Assumptions: Pre-period violations bound the magnitude of post-period violations (relative-magnitude family); Post-treatment effects are constant across event time (relax via alternative parameter families in Rambachan-Roth 2023 Section 3). Pre-conditions: result has at least one pre-period and one post-period coefficient; result carries the variance-covariance matrix of those coefficients. Failure modes: Breakdown Mbar < 1.0 (small parallel-trends violation overturns the sign) -> The result is fragile to plausible pretrends violations; report the breakdown alongside the point estimate; Confidence set is the entire real line (Mbar grid too coarse) -> Re-run with a finer grid (n_grid=50+) or restrict Mbar to a tighter interval. Alternatives: sp.honest_did, sp.pretrends_test, sp.breakdown_m. Typical minimum N: 50.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`Mbar`	No	Grid of relative-magnitude bounds; default is np.linspace(0, 2, n_grid)
`alpha`	No	Significance level for confidence intervals and tests.
`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	Identification method	C-LF
`n_grid`	No	Mbar grid size when Mbar=None
`result`	Yes	Event-study or DiD result with full pre/post coefficients
`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.

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

A4.6/5.0

Behavior5/5

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

Watch annotations only mark readOnlyHint=true, which this description never contradicts. Beyond that, it transparently discloses assumptions, failure modes (fragile when Mbar<1.0; grid too coarse), and interpretation guidance. This is far more behavioral context than the annotations provide and directly helps the agent interpret outputs correctly.

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

Conciseness4/5

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

The description is dense but well-organized: main function and outputs first, then assumptions, pre-conditions, failure modes, and alternatives in labeled segments. No sentence is purely filler, though the length is considerable. It earns its size for a complex sensitivity-analysis tool.

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

Completeness5/5

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

The description covers the method, output interpretation, assumptions, pre-conditions, failure modes, alternatives, and even a typical minimum N. The companion output schema likely carries return details, so nothing an agent needs for correct invocation or interpretation appears to be missing.

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

Parameters4/5

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

The input schema already has 100% parameter description coverage, establishing a baseline of 3. The description adds extra meaning for key parameters: it explains the Mbar parametrization ('relative magnitude of post-period violation versus worst pre-period one'), and the failure-mode guidance gives actionable meaning to n_grid (finer grid, n_grid=50+). It also clarifies what the 'result' must contain (pre/post coefficients plus variance-covariance matrix), which is not fully specified in the schema.

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

Purpose5/5

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

The description states a specific verb ('computes'), a precise resource (Rambachan-Roth honest-DiD sensitivity), and a well-defined output (breakdown Mbar and confidence sets). It also names alternatives (sp.honest_did, sp.pretrends_test, sp.breakdown_m), so an agent can distinguish this from related tools. This is a clear, differentiated statement of what the tool does.

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

Usage Guidelines4/5

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

The description provides concrete pre-conditions (result has pre/post coefficients with a variance-covariance matrix), assumptions (relative-magnitude family, constant effects), and failure modes that clarify when results are useful. It lists alternatives but does not give explicit decision rules for choosing among them, so it stops slightly short of full when-to-use/when-not-to-use guidance.

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

Deploy Server