Factorguide

bwana7/factorguide

Connects to FactorGuide's decision intelligence API for analyzing coupling structures in systems with interacting variables. Exposes tools for sending covariance or correlation matrices and receiving zone classifications that tell you which variable pairs can be safely factorized and which couplings must be preserved. Returns concrete recommendations with IC scores, mutual information costs, and risk predictions calibrated on validated datasets. Useful when an agent needs to simplify a complex system without losing critical dependencies, like deciding which variables to treat as independent in a model or spotting load-bearing correlations before refactoring. Accepts only summary statistics, never raw data. Pricing starts at 15 free queries, then pay per query with USDC on Base.

tools: 7
transport: HTTP
last updated: Apr 3, 2026

View on GitHub

Install

From registry metadata

$ npx add-mcp https://factorguide.io/mcp --name factorguide

Run in your terminal. add-mcp asks which agents to add it to (Claude Code, Codex, Cursor and more). Add -g to install it for every project.

Review the command, arguments and environment values before installing. MCP servers run with your local permissions.

Other ways

Claude Code

$ claude mcp add --transport http factorguide https://factorguide.io/mcp

Run in your terminal. Add --scope user to make it available in every project.

Tools

Verified live · 7 tools

Verified live against the running server on Nov 6, 2026.

factorguide_navigateMap the factorization terrain of your model. Send coupling structure (precision matrix preferred for n>2; covariance matrix recommended if sign or CC information is needed) and receive a block-diagonal strategy with calibrated risk prediction. Answers: 'How should I factorize,...13 params

Map the factorization terrain of your model. Send coupling structure (precision matrix preferred for n>2; covariance matrix recommended if sign or CC information is needed) and receive a block-diagonal strategy with calibrated risk prediction. Answers: 'How should I factorize,...

Parameters* required

accuracy_targetnumber

default: 2

compute_budgetvalue

default: minimize

cost_modelvalue

default: cubic

coupling*value

distribution_diagnosticsvalue

encoding_labelvalue

model_classvalue

default: unknown

report_marginal_icboolean

default: false

report_sign_detectabilityboolean

default: false

sample_size*integer

synergy_checkboolean

default: false

task_typevalue

default: inference

variable_namesvalue

factorguide_diagnoseQuick single-pair diagnostic. IC with risk prediction for both model classes. Include variances for sign detectability. Requires X-Wallet header with your EVM wallet address (0x...). First 5 queries are free trial.6 params

Quick single-pair diagnostic. IC with risk prediction for both model classes. Include variances for sign detectability. Requires X-Wallet header with your EVM wallet address (0x...). First 5 queries are free trial.

Parameters* required

coupling_value*number

IC or coupling value

i*string

First variable name

j*string

Second variable name

sample_size*integer

variance_inumber

variance_jnumber

factorguide_submit_paymentSubmit payment proof after sending stablecoins to a FactorGuide wallet address. For x402: provide tx_hash and chain. For MPP: use in-band Authorization header instead — no separate submission needed.2 params

Submit payment proof after sending stablecoins to a FactorGuide wallet address. For x402: provide tx_hash and chain. For MPP: use in-band Authorization header instead — no separate submission needed.

Parameters* required

chain*string

Chain identifier, e.g. 'eip155:8453' or 'tempo:4217'

tx_hash*string

On-chain transaction hash

factorguide_report_outcomeComplete the prediction loop — report inference diagnostics so future predictions improve. After running the approach FactorGuide recommended, return your ESS ratio, PSIS-khat, or log-likelihood gap. Zero additional computation required. Does not consume a query allocation.8 params

Complete the prediction loop — report inference diagnostics so future predictions improve. After running the approach FactorGuide recommended, return your ESS ratio, PSIS-khat, or log-likelihood gap. Zero additional computation required. Does not consume a query allocation.

Parameters* required

actual_mse_ratiovalue

approach_taken*value

ess_ratiovalue

log_lik_gapvalue

n_replicationsvalue

prediction_hash*string

psis_khatvalue

runtime_secondsvalue

factorguide_explainPlain-language explanation of a previous navigate response, including wave mechanics grounding for observational cost guidance. Requires a prediction_hash from a prior factorguide_navigate call. Consumes 1 query allocation. Available for starter and professional tiers. Require...1 params

Plain-language explanation of a previous navigate response, including wave mechanics grounding for observational cost guidance. Requires a prediction_hash from a prior factorguide_navigate call. Consumes 1 query allocation. Available for starter and professional tiers. Require...

Parameters* required

prediction_hash*string

prediction_hash from a previous navigate response

factorguide_regime_detectDetect coupling regime changes in time series via windowed IC. Specification pending — v1.1 target.

Detect coupling regime changes in time series via windowed IC. Specification pending — v1.1 target.

No parameters. Call it with no arguments.

factorguide_synergy_detectDetect hidden synergistic structure via Walsh-Hadamard spectral analysis. Accepts pre-computed Walsh coefficients — agent performs the transform locally and sends only the spectral summary. Specification pending — v1.1 target.5 params

Detect hidden synergistic structure via Walsh-Hadamard spectral analysis. Accepts pre-computed Walsh coefficients — agent performs the transform locally and sends only the spectral summary. Specification pending — v1.1 target.

Parameters* required

ic_matrix_refstring

n_samplesinteger

n_variablesinteger

transform_methodstring

one of exact · sampled

walsh_coefficientsobject

Readme

FactorGuide

Decision intelligence for AI agents. Send a coupling matrix — get zone classifications, optimal factorization strategy, and calibrated risk predictions.

Every system with interacting variables has a coupling structure. When an agent simplifies that system by treating variables as independent, it pays an information cost. FactorGuide quantifies that cost exactly.

Quick Start

MCP (Model Context Protocol)

FactorGuide is listed on the Official MCP Registry. Any MCP-compatible agent discovers tools automatically:

mcp_endpoint: https://factorguide.io/mcp

HTTP REST

POST https://factorguide.io/navigate
POST https://factorguide.io/diagnose
POST https://factorguide.io/explain
POST https://factorguide.io/report_outcome

Agent Discovery

GET https://factorguide.io/llms.txt
GET https://factorguide.io/openapi.json

Example

Send a 3×3 correlation matrix:

{
  "coupling": {
    "covariance_matrix": [
      [1.00, 0.72, 0.05],
      [0.72, 1.00, 0.48],
      [0.05, 0.48, 1.00]
    ]
  },
  "sample_size": 500,
  "model_class": "constitutive"
}

View the full README on GitHub

Details

Repository: github.com/bwana7/factorguide
Category: Communication & Messaging
Transport: HTTP
Registry: MCP registry, active
Tools verified: Nov 6, 2026

Related Communication & Messaging MCP servers

View all

Bizfilecloudcorprecords/bizfileVerify businesses against public registries and check email headers, with signed receipts.0stars

AY Inboxayinbox/ayinbox-mcpCold email from your agent: contact lists, campaigns, campaign brains, reply drafts and a CRM.0stars

Blackwallbluetieroperations-create/blackwall-mcpAPI-key pre-action risk gate: any irreversible agent action (money, SQL, delete, email).0stars

Mcpbounceprotect/bounceprotect-mcpEmail validation and SMTP verification for Claude Desktop, Cursor, and Claude Code.0stars

GlucoBeatdmbejar/glucobeat-mcpLog glucose, meals, exercise and sleep in your GlucoBeat diary by chat, and look up your own data0stars

MyAITwinlutolabs/mytwin-mcpA personal RAG database you build from chat, so AI creates work that sounds like you.0stars