Black-Swan-Causal-Labs/dagstudio-mcp

github
github JavaScript analyzed · update queued

MCP server for the DAG Studio causal-inference engine: backdoor analysis, adjustment sets, bias simulation, effect-modification classification. Validated against dagitty.

maintainer
Black-Swan-Causal-Labs
licence
Apache-2.0
first seen
2026-07-19
last seen
2026-11-12
releases · 30d
0
github stars
0 reported
short id

No releases · last push 3d ago

No releases or dependencies recorded yet.

A minimal

what we found

Its 9 tools appear to reach network, inferred from source code and tool definitions rather than from observed behaviour.

full security breakdown →

tools

9
  • analyze_dag Backdoor paths, minimal sufficient adjustment sets, identifiability
  • check_overadjustment Detect adjustment for mediators, colliders, and descendants of exposure
  • classify_effect_modification Classify effect-modifier structure (direct, indirect, proxy, common-cause, pure interaction)
  • compute_bias Empirical bias of an adjustment strategy against the simulated truth
  • generate_code R / Python analysis code for a DAG, plus a one-click DAG Studio URL
  • get_canonical_example Canonical teaching DAGs (confounding, M-bias, frontdoor, and others)
  • parse_dagitty Parse dagitty DSL (raw or R-wrapped) into the structured DAG model
  • simulate_data Simulate data from a DAG under user-specified structural coefficients
  • validate_engine Run the full canonical validation suite and report engine version

Black-Swan-Causal-Labs/dagstudio-mcp is an MCP server distributed on github, maintained by Black-Swan-Causal-Labs, tracked here since July 2026. It has shipped 0 releases and exposes 9 tools. Tools include analyze_dag, check_overadjustment, classify_effect_modification, compute_bias, generate_code, get_canonical_example, and 3 more. Its composite risk grade is A — an inferred review prompt computed from observed signals, not a verdict.