B4 Index MCP · Ben Roberts AI

Operational

Unclaimed — claim this pagedata Streamable HTTP2025-06-18 API key expected

https://b4-index.vercel.app/mcp

Badge

100%

Availability · 30 days · 357 checks

2099 ms

Latency (last observation)

rev 3

Contract

10days

Schema stable

Incidents · 30 days

Aug 6, 2026

Observed since

Availability — last 30 days

No failures Some failures Mostly failing Auth required No observations

Last observed 39 min ago by MCP Tools Edge Scanner · scheduled every 120 min

Non-breaking contract changes (9) Non-breaking Sep 7, 2026 — 5:05 PM

5 non-breaking changes. Tool b4_audit now declares an output schema.

VerifiedTool b4_audit now declares an output schema.

InterpretationA new guarantee for consumers.

Raw event details

{
  "beforeRev": 2,
  "afterRev": 3,
  "changeCount": 9
}

View full diff Observed by MCP Tools Edge Scanner

Descriptions or metadata changed Informational Sep 7, 2026 — 3:05 PM

5 description/metadata changes; the functional contract is unchanged (hash sha256:e9992024d6bda77f…).

VerifiedDescriptions or metadata for b4_audit changed.

InterpretationNo structural change. Wording changes can still influence model tool selection.

Path	Change	Classification	Detail
tools.b4_audit	description changed	Informational	Descriptions or metadata for `b4_audit` changed. No structural change. Wording changes can still influence model tool selection.
tools.b4_browse	description changed	Informational	Descriptions or metadata for `b4_browse` changed. No structural change. Wording changes can still influence model tool selection.
tools.b4_compare	description changed	Informational	Descriptions or metadata for `b4_compare` changed. No structural change. Wording changes can still influence model tool selection.
tools.b4_recommend	description changed	Informational	Descriptions or metadata for `b4_recommend` changed. No structural change. Wording changes can still influence model tool selection.
tools.b4_score	description changed	Informational	Descriptions or metadata for `b4_score` changed. No structural change. Wording changes can still influence model tool selection.

Observed by MCP Tools Edge Scanner

Descriptions or metadata changed Informational Aug 12, 2026 — 4:50 AM

3 description/metadata changes; the functional contract is unchanged (hash sha256:e9992024d6bda77f…).

VerifiedDescriptions or metadata for b4_audit changed.

InterpretationNo structural change. Wording changes can still influence model tool selection.

Path	Change	Classification	Detail
tools.b4_audit	description changed	Informational	Descriptions or metadata for `b4_audit` changed. No structural change. Wording changes can still influence model tool selection.
tools.b4_compare	description changed	Informational	Descriptions or metadata for `b4_compare` changed. No structural change. Wording changes can still influence model tool selection.
tools.b4_recommend	description changed	Informational	Descriptions or metadata for `b4_recommend` changed. No structural change. Wording changes can still influence model tool selection.

Observed by MCP Tools Edge Scanner

Non-breaking contract changes (10) Non-breaking Aug 9, 2026 — 2:20 PM

5 non-breaking changes. Optional parameter org was added at b4_audit.inputSchema.

VerifiedOptional parameter org was added at b4_audit.inputSchema.

InterpretationExisting calls are unaffected.

Raw event details

{
  "beforeRev": 1,
  "afterRev": 2,
  "changeCount": 10
}

View full diff Observed by MCP Tools Edge Scanner

Baseline captured Aug 6, 2026 — 4:55 AM

5 tools, 0 resources, 0 prompts observed. Future observations will identify outages, recoveries, protocol changes, and contract drift from this state.

VerifiedCanonical contract hash sha256:c07d2ab1b77e9d24… computed from the normalized capability inventory.

Raw event details

{
  "contractHash": "sha256:c07d2ab1b77e9d24dbcb1a1576ab822d04fb126c88a8642466ba115ef9fd2886"
}

Observed by MCP Tools Edge Scanner

Advertised capabilities

Tools · 5 Resources Prompts Logging Completions Experimental

Tools 5

b4_browse 6 params Search and filter the B4 Index's 1,600+ independently scored software categories. Browse by keyword, domain, q…

Search and filter the B4 Index's 1,600+ independently scored software categories. Browse by keyword, domain, quadrant, or industry. When filtering by industry, returns all vertical categories for that industry PLUS all horizontal categories (which apply to every industry). Each row carries its banded verdict — primary, confidence word, and a near-call flag — and the quadrant filter matches the verdict at whichever lens you are reading. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to "small", "medium" (the default) or "large" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [Needs a B4 plan: browse and score come with B4 Web.]

Input schema

Parameter	Type	Description & constraints
org	enum(3)	Org-maturity lens: "small" (no dedicated engineering), "medium" (default — some AI capability), "large" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile. one of: "small", "medium", "large" · default: "medium"
query	string	Search term to match against category names, vendors, domains, and rationales maxLen: 200
domain	string	Filter by domain (e.g., 'Marketing Technology', 'CRM & Sales') maxLen: 120
quadrant	enum(4)	Filter by quadrant one of: "BUILD", "BUY", "BRIDGE", "BEWARE"
industry	string	Filter by industry group. Returns matching vertical categories + all horizontal categories. Options: Healthcare, Financial Services, Construction & Real Estate, Education, Energy & Utilities, Government, Automotive, Agriculture, Transportation & Logistics, Media & Entertainment, Legal, Professional Services, Nonprofits & Associations, Manufacturing, Retail & Commerce, Hospitality & Food Service, Telecom maxLen: 120
limit	integer	Max results to return (default 20, max 100) min: 1 · max: 100 · default: 20

Output schema

Parameter	Type	Description & constraints
resolved*	boolean	false when nothing matched; the payload then carries no verdict
category	string \| null	—
domain	string \| null	—
lens	string \| null	—
verdict	object	banded verdict: primary, confidence, nearCall, distribution
scores	object	the five v3.2 dimension scores, 1-5
axes	object	strategicDifferentiation (X) and aiFeasibility (Y)
urgency	object	—
rationale	string \| null	—
aiReplacement	object	—
vendors	object	—
vendorCount	number	—
vendorsWithheld	number	—
evidence	object	sources and research trail; present only with includeEvidence
evidenceAvailable	boolean	—
evidenceHint	string	—
guidance	object	how to use this result: meaning, action, what not to claim, confidence handling

Raw schema JSON

b4_browse — raw

{
  "name": "b4_browse",
  "description": "Search and filter the B4 Index&#39;s 1,600+ independently scored software categories. Browse by keyword, domain, quadrant, or industry. When filtering by industry, returns all vertical categories for that industry PLUS all horizontal categories (which apply to every industry). Each row carries its banded verdict — primary, confidence word, and a near-call flag — and the quadrant filter matches the verdict at whichever lens you are reading. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to \"small\", \"medium\" (the default) or \"large\" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [Needs a B4 plan: browse and score come with B4 Web.]",
  "inputSchema": {
    "type": "object",
    "properties": {
      "org": {
        "type": "string",
        "enum": [
          "small",
          "medium",
          "large"
        ],
        "default": "medium",
        "description": "Org-maturity lens: \"small\" (no dedicated engineering), \"medium\" (default — some AI capability), \"large\" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile."
      },
      "query": {
        "type": "string",
        "maxLength": 200,
        "description": "Search term to match against category names, vendors, domains, and rationales"
      },
      "domain": {
        "type": "string",
        "maxLength": 120,
        "description": "Filter by domain (e.g., &#39;Marketing Technology&#39;, &#39;CRM & Sales&#39;)"
      },
      "quadrant": {
        "type": "string",
        "enum": [
          "BUILD",
          "BUY",
          "BRIDGE",
          "BEWARE"
        ],
        "description": "Filter by quadrant"
      },
      "industry": {
        "type": "string",
        "maxLength": 120,
        "description": "Filter by industry group. Returns matching vertical categories + all horizontal categories. Options: Healthcare, Financial Services, Construction & Real Estate, Education, Energy & Utilities, Government, Automotive, Agriculture, Transportation & Logistics, Media & Entertainment, Legal, Professional Services, Nonprofits & Associations, Manufacturing, Retail & Commerce, Hospitality & Food Service, Telecom"
      },
      "limit": {
        "type": "integer",
        "minimum": 1,
        "maximum": 100,
        "default": 20,
        "description": "Max results to return (default 20, max 100)"
      }
    },
    "additionalProperties": false,
    "$schema": "http://json-schema.org/draft-07/schema#"
  },
  "outputSchema": {
    "type": "object",
    "properties": {
      "resolved": {
        "type": "boolean",
        "description": "false when nothing matched; the payload then carries no verdict"
      },
      "category": {
        "type": [
          "string",
          "null"
        ]
      },
      "domain": {
        "type": [
          "string",
          "null"
        ]
      },
      "lens": {
        "type": [
          "string",
          "null"
        ]
      },
      "verdict": {
        "description": "banded verdict: primary, confidence, nearCall, distribution"
      },
      "scores": {
        "description": "the five v3.2 dimension scores, 1-5"
      },
      "axes": {
        "description": "strategicDifferentiation (X) and aiFeasibility (Y)"
      },
      "urgency": {},
      "rationale": {
        "type": [
          "string",
          "null"
        ]
      },
      "aiReplacement": {},
      "vendors": {},
      "vendorCount": {
        "type": "number"
      },
      "vendorsWithheld": {
        "type": "number"
      },
      "evidence": {
        "description": "sources and research trail; present only with includeEvidence"
      },
      "evidenceAvailable": {
        "type": "boolean"
      },
      "evidenceHint": {
        "type": "string"
      },
      "guidance": {
        "description": "how to use this result: meaning, action, what not to claim, confidence handling"
      }
    },
    "required": [
      "resolved"
    ],
    "additionalProperties": true,
    "$schema": "http://json-schema.org/draft-07/schema#"
  }
}

b4_score 4 params Score a software category using the B4 Index. Provide a known category name to get pre-computed scores, or pro…

Score a software category using the B4 Index. Provide a known category name to get pre-computed scores, or provide raw dimension scores (1-5 each) for a custom evaluation. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Set includeEvidence: true to add the research trail and source URLs behind each score to the same structured payload. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to "small", "medium" (the default) or "large" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [Needs a B4 plan: browse and score come with B4 Web.]

Input schema

Parameter	Type	Description & constraints
org	enum(3)	Org-maturity lens: "small" (no dedicated engineering), "medium" (default — some AI capability), "large" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile. one of: "small", "medium", "large" · default: "medium"
category	string	Name of a known B4 category (e.g., 'Expense Management', 'CRM') maxLen: 120
includeEvidence	boolean	Include the full evidence trail and source URLs behind each dimension score. Off by default so the initial result stays concise; set true for deep verification. default: false
scores	object	Custom dimension scores for a tool not in the database
scores.specificity*	number	1-5: How company-specific is the need? min: 1 · max: 5
scores.aiFeasibility*	number	1-5: How feasible is AI replacement? min: 1 · max: 5
scores.vendorValue*	number	1-5: How much vendor value are you NOT using? (higher = more waste) min: 1 · max: 5
scores.strategicControl*	number	1-5: How strategically important is owning this? min: 1 · max: 5
scores.costTrajectory*	number	1-5: How much is build cost beating vendor cost? min: 1 · max: 5

Output schema

Parameter	Type	Description & constraints
resolved*	boolean	false when nothing matched; the payload then carries no verdict
category	string \| null	—
domain	string \| null	—
lens	string \| null	—
verdict	object	banded verdict: primary, confidence, nearCall, distribution
scores	object	the five v3.2 dimension scores, 1-5
axes	object	strategicDifferentiation (X) and aiFeasibility (Y)
urgency	object	—
rationale	string \| null	—
aiReplacement	object	—
vendors	object	—
vendorCount	number	—
vendorsWithheld	number	—
evidence	object	sources and research trail; present only with includeEvidence
evidenceAvailable	boolean	—
evidenceHint	string	—
guidance	object	how to use this result: meaning, action, what not to claim, confidence handling

Raw schema JSON

b4_score — raw

{
  "name": "b4_score",
  "description": "Score a software category using the B4 Index. Provide a known category name to get pre-computed scores, or provide raw dimension scores (1-5 each) for a custom evaluation. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Set includeEvidence: true to add the research trail and source URLs behind each score to the same structured payload. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to \"small\", \"medium\" (the default) or \"large\" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [Needs a B4 plan: browse and score come with B4 Web.]",
  "inputSchema": {
    "type": "object",
    "properties": {
      "org": {
        "type": "string",
        "enum": [
          "small",
          "medium",
          "large"
        ],
        "default": "medium",
        "description": "Org-maturity lens: \"small\" (no dedicated engineering), \"medium\" (default — some AI capability), \"large\" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile."
      },
      "category": {
        "type": "string",
        "maxLength": 120,
        "description": "Name of a known B4 category (e.g., &#39;Expense Management&#39;, &#39;CRM&#39;)"
      },
      "includeEvidence": {
        "type": "boolean",
        "default": false,
        "description": "Include the full evidence trail and source URLs behind each dimension score. Off by default so the initial result stays concise; set true for deep verification."
      },
      "scores": {
        "type": "object",
        "properties": {
          "specificity": {
            "type": "number",
            "minimum": 1,
            "maximum": 5,
            "description": "1-5: How company-specific is the need?"
          },
          "aiFeasibility": {
            "type": "number",
            "minimum": 1,
            "maximum": 5,
            "description": "1-5: How feasible is AI replacement?"
          },
          "vendorValue": {
            "type": "number",
            "minimum": 1,
            "maximum": 5,
            "description": "1-5: How much vendor value are you NOT using? (higher = more waste)"
          },
          "strategicControl": {
            "type": "number",
            "minimum": 1,
            "maximum": 5,
            "description": "1-5: How strategically important is owning this?"
          },
          "costTrajectory": {
            "type": "number",
            "minimum": 1,
            "maximum": 5,
            "description": "1-5: How much is build cost beating vendor cost?"
          }
        },
        "required": [
          "specificity",
          "aiFeasibility",
          "vendorValue",
          "strategicControl",
          "costTrajectory"
        ],
        "additionalProperties": false,
        "description": "Custom dimension scores for a tool not in the database"
      }
    },
    "additionalProperties": false,
    "$schema": "http://json-schema.org/draft-07/schema#"
  },
  "outputSchema": {
    "type": "object",
    "properties": {
      "resolved": {
        "type": "boolean",
        "description": "false when nothing matched; the payload then carries no verdict"
      },
      "category": {
        "type": [
          "string",
          "null"
        ]
      },
      "domain": {
        "type": [
          "string",
          "null"
        ]
      },
      "lens": {
        "type": [
          "string",
          "null"
        ]
      },
      "verdict": {
        "description": "banded verdict: primary, confidence, nearCall, distribution"
      },
      "scores": {
        "description": "the five v3.2 dimension scores, 1-5"
      },
      "axes": {
        "description": "strategicDifferentiation (X) and aiFeasibility (Y)"
      },
      "urgency": {},
      "rationale": {
        "type": [
          "string",
          "null"
        ]
      },
      "aiReplacement": {},
      "vendors": {},
      "vendorCount": {
        "type": "number"
      },
      "vendorsWithheld": {
        "type": "number"
      },
      "evidence": {
        "description": "sources and research trail; present only with includeEvidence"
      },
      "evidenceAvailable": {
        "type": "boolean"
      },
      "evidenceHint": {
        "type": "string"
      },
      "guidance": {
        "description": "how to use this result: meaning, action, what not to claim, confidence handling"
      }
    },
    "required": [
      "resolved"
    ],
    "additionalProperties": true,
    "$schema": "http://json-schema.org/draft-07/schema#"
  }
}

b4_audit 2 params · 1 required Analyze a software stack against the B4 Index. Provide a list of tool/category names, and get per-tool banded …

Analyze a software stack against the B4 Index. Provide a list of tool/category names, and get per-tool banded verdicts plus a portfolio verdict distribution. Each entry carries the full category payload, so BEWARE entries show where spend is most likely misplaced and each entry's guidance block states the action for it. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to "small", "medium" (the default) or "large" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [B4 Agent tool. Browse and score come with B4 Web; this one needs Agent.]

Input schema

Parameter	Type	Description & constraints
org	enum(3)	Org-maturity lens: "small" (no dedicated engineering), "medium" (default — some AI capability), "large" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile. one of: "small", "medium", "large" · default: "medium"
tools*	array<string>	List of software tool or category names to audit (e.g., ['Salesforce', 'Slack', 'Expense Management']). Max 100 per call.

Output schema

Parameter	Type	Description & constraints
resolved*	boolean	false when nothing matched; the payload then carries no verdict
category	string \| null	—
domain	string \| null	—
lens	string \| null	—
verdict	object	banded verdict: primary, confidence, nearCall, distribution
scores	object	the five v3.2 dimension scores, 1-5
axes	object	strategicDifferentiation (X) and aiFeasibility (Y)
urgency	object	—
rationale	string \| null	—
aiReplacement	object	—
vendors	object	—
vendorCount	number	—
vendorsWithheld	number	—
evidence	object	sources and research trail; present only with includeEvidence
evidenceAvailable	boolean	—
evidenceHint	string	—
guidance	object	how to use this result: meaning, action, what not to claim, confidence handling

Raw schema JSON

b4_audit — raw

{
  "name": "b4_audit",
  "description": "Analyze a software stack against the B4 Index. Provide a list of tool/category names, and get per-tool banded verdicts plus a portfolio verdict distribution. Each entry carries the full category payload, so BEWARE entries show where spend is most likely misplaced and each entry&#39;s guidance block states the action for it. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to \"small\", \"medium\" (the default) or \"large\" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [B4 Agent tool. Browse and score come with B4 Web; this one needs Agent.]",
  "inputSchema": {
    "type": "object",
    "properties": {
      "org": {
        "type": "string",
        "enum": [
          "small",
          "medium",
          "large"
        ],
        "default": "medium",
        "description": "Org-maturity lens: \"small\" (no dedicated engineering), \"medium\" (default — some AI capability), \"large\" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile."
      },
      "tools": {
        "type": "array",
        "items": {
          "type": "string",
          "maxLength": 120
        },
        "minItems": 1,
        "maxItems": 100,
        "description": "List of software tool or category names to audit (e.g., [&#39;Salesforce&#39;, &#39;Slack&#39;, &#39;Expense Management&#39;]). Max 100 per call."
      }
    },
    "required": [
      "tools"
    ],
    "additionalProperties": false,
    "$schema": "http://json-schema.org/draft-07/schema#"
  },
  "outputSchema": {
    "type": "object",
    "properties": {
      "resolved": {
        "type": "boolean",
        "description": "false when nothing matched; the payload then carries no verdict"
      },
      "category": {
        "type": [
          "string",
          "null"
        ]
      },
      "domain": {
        "type": [
          "string",
          "null"
        ]
      },
      "lens": {
        "type": [
          "string",
          "null"
        ]
      },
      "verdict": {
        "description": "banded verdict: primary, confidence, nearCall, distribution"
      },
      "scores": {
        "description": "the five v3.2 dimension scores, 1-5"
      },
      "axes": {
        "description": "strategicDifferentiation (X) and aiFeasibility (Y)"
      },
      "urgency": {},
      "rationale": {
        "type": [
          "string",
          "null"
        ]
      },
      "aiReplacement": {},
      "vendors": {},
      "vendorCount": {
        "type": "number"
      },
      "vendorsWithheld": {
        "type": "number"
      },
      "evidence": {
        "description": "sources and research trail; present only with includeEvidence"
      },
      "evidenceAvailable": {
        "type": "boolean"
      },
      "evidenceHint": {
        "type": "string"
      },
      "guidance": {
        "description": "how to use this result: meaning, action, what not to claim, confidence handling"
      }
    },
    "required": [
      "resolved"
    ],
    "additionalProperties": true,
    "$schema": "http://json-schema.org/draft-07/schema#"
  }
}

b4_compare 2 params · 1 required Compare build vs buy for a specific software category. Returns side-by-side analysis with the category's bande…

Compare build vs buy for a specific software category. Returns side-by-side analysis with the category's banded verdict, scores, vendor options, AI replacement approach, and action steps for each path. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to "small", "medium" (the default) or "large" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [B4 Agent tool. Browse and score come with B4 Web; this one needs Agent.]

Input schema

Parameter	Type	Description & constraints
org	enum(3)	Org-maturity lens: "small" (no dedicated engineering), "medium" (default — some AI capability), "large" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile. one of: "small", "medium", "large" · default: "medium"
category*	string	Name of the software category to compare (e.g., 'Email Marketing', 'CRM') maxLen: 120

Output schema

Parameter	Type	Description & constraints
resolved*	boolean	false when nothing matched; the payload then carries no verdict
category	string \| null	—
domain	string \| null	—
lens	string \| null	—
verdict	object	banded verdict: primary, confidence, nearCall, distribution
scores	object	the five v3.2 dimension scores, 1-5
axes	object	strategicDifferentiation (X) and aiFeasibility (Y)
urgency	object	—
rationale	string \| null	—
aiReplacement	object	—
vendors	object	—
vendorCount	number	—
vendorsWithheld	number	—
evidence	object	sources and research trail; present only with includeEvidence
evidenceAvailable	boolean	—
evidenceHint	string	—
guidance	object	how to use this result: meaning, action, what not to claim, confidence handling

Raw schema JSON

b4_compare — raw

{
  "name": "b4_compare",
  "description": "Compare build vs buy for a specific software category. Returns side-by-side analysis with the category&#39;s banded verdict, scores, vendor options, AI replacement approach, and action steps for each path. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to \"small\", \"medium\" (the default) or \"large\" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [B4 Agent tool. Browse and score come with B4 Web; this one needs Agent.]",
  "inputSchema": {
    "type": "object",
    "properties": {
      "org": {
        "type": "string",
        "enum": [
          "small",
          "medium",
          "large"
        ],
        "default": "medium",
        "description": "Org-maturity lens: \"small\" (no dedicated engineering), \"medium\" (default — some AI capability), \"large\" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile."
      },
      "category": {
        "type": "string",
        "maxLength": 120,
        "description": "Name of the software category to compare (e.g., &#39;Email Marketing&#39;, &#39;CRM&#39;)"
      }
    },
    "required": [
      "category"
    ],
    "additionalProperties": false,
    "$schema": "http://json-schema.org/draft-07/schema#"
  },
  "outputSchema": {
    "type": "object",
    "properties": {
      "resolved": {
        "type": "boolean",
        "description": "false when nothing matched; the payload then carries no verdict"
      },
      "category": {
        "type": [
          "string",
          "null"
        ]
      },
      "domain": {
        "type": [
          "string",
          "null"
        ]
      },
      "lens": {
        "type": [
          "string",
          "null"
        ]
      },
      "verdict": {
        "description": "banded verdict: primary, confidence, nearCall, distribution"
      },
      "scores": {
        "description": "the five v3.2 dimension scores, 1-5"
      },
      "axes": {
        "description": "strategicDifferentiation (X) and aiFeasibility (Y)"
      },
      "urgency": {},
      "rationale": {
        "type": [
          "string",
          "null"
        ]
      },
      "aiReplacement": {},
      "vendors": {},
      "vendorCount": {
        "type": "number"
      },
      "vendorsWithheld": {
        "type": "number"
      },
      "evidence": {
        "description": "sources and research trail; present only with includeEvidence"
      },
      "evidenceAvailable": {
        "type": "boolean"
      },
      "evidenceHint": {
        "type": "string"
      },
      "guidance": {
        "description": "how to use this result: meaning, action, what not to claim, confidence handling"
      }
    },
    "required": [
      "resolved"
    ],
    "additionalProperties": true,
    "$schema": "http://json-schema.org/draft-07/schema#"
  }
}

b4_recommend 2 params · 1 required Get B4 Index recommendations from a natural language description of a software need or business context. Match…

Get B4 Index recommendations from a natural language description of a software need or business context. Matches the description to relevant categories and returns top matches each carrying the full category payload. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to "small", "medium" (the default) or "large" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [B4 Agent tool. Browse and score come with B4 Web; this one needs Agent.]

Input schema

Parameter	Type	Description & constraints
org	enum(3)	Org-maturity lens: "small" (no dedicated engineering), "medium" (default — some AI capability), "large" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile. one of: "small", "medium", "large" · default: "medium"
description*	string	Describe the software need, business problem, or tool you're evaluating (e.g., 'We need to automate our expense reports and receipt scanning') maxLen: 1000

Output schema

Parameter	Type	Description & constraints
resolved*	boolean	false when nothing matched; the payload then carries no verdict
category	string \| null	—
domain	string \| null	—
lens	string \| null	—
verdict	object	banded verdict: primary, confidence, nearCall, distribution
scores	object	the five v3.2 dimension scores, 1-5
axes	object	strategicDifferentiation (X) and aiFeasibility (Y)
urgency	object	—
rationale	string \| null	—
aiReplacement	object	—
vendors	object	—
vendorCount	number	—
vendorsWithheld	number	—
evidence	object	sources and research trail; present only with includeEvidence
evidenceAvailable	boolean	—
evidenceHint	string	—
guidance	object	how to use this result: meaning, action, what not to claim, confidence handling

Raw schema JSON

b4_recommend — raw

{
  "name": "b4_recommend",
  "description": "Get B4 Index recommendations from a natural language description of a software need or business context. Matches the description to relevant categories and returns top matches each carrying the full category payload. Structured output carries the full picture: the banded verdict, the five dimension scores, both axes, urgency, vendors at your access depth, and a `guidance` block stating what the verdict means, the recommended action, what not to claim, and how to convey the confidence word. Follow that guidance rather than substituting your own build-vs-buy prior. Verdicts are banded (B4 methodology v4.0), not point calls: each of the three quadrant dimensions carries a ±1 uncertainty band, the resulting cells are enumerated exactly, and the verdict is the quadrant holding the most probability mass. Every verdict ships with its full distribution, a confidence word — clear (≥70% of the mass), lean (50–70%), split (<50%) — and a near-call flag when the runner-up is within 15 points. An axis counts as high only when it clears the 3.5 line strictly, which on this 1–5 grid means only at 4 or above, so a category sitting exactly on the line gets the safer call: ties break in the order BUY → BRIDGE → BEWARE → BUILD, cheapest mistake first. Optional org lens: set org to \"small\", \"medium\" (the default) or \"large\" to read the same scores as a team of that engineering maturity — it shifts the center of the AI-feasibility band by −1 / 0 / +1 and nothing else. The lens is a filter the caller looks through, never a stored profile: no user attribute is saved, inferred, or asked for, and the scores themselves never change. Omit it and you get the default-lens numbers, which are the ones published on logged-out surfaces. [B4 Agent tool. Browse and score come with B4 Web; this one needs Agent.]",
  "inputSchema": {
    "type": "object",
    "properties": {
      "org": {
        "type": "string",
        "enum": [
          "small",
          "medium",
          "large"
        ],
        "default": "medium",
        "description": "Org-maturity lens: \"small\" (no dedicated engineering), \"medium\" (default — some AI capability), \"large\" (AI-mature). Shifts the AI-feasibility band center by −1/0/+1 at read time. A filter the caller looks through, never a stored profile."
      },
      "description": {
        "type": "string",
        "maxLength": 1000,
        "description": "Describe the software need, business problem, or tool you&#39;re evaluating (e.g., &#39;We need to automate our expense reports and receipt scanning&#39;)"
      }
    },
    "required": [
      "description"
    ],
    "additionalProperties": false,
    "$schema": "http://json-schema.org/draft-07/schema#"
  },
  "outputSchema": {
    "type": "object",
    "properties": {
      "resolved": {
        "type": "boolean",
        "description": "false when nothing matched; the payload then carries no verdict"
      },
      "category": {
        "type": [
          "string",
          "null"
        ]
      },
      "domain": {
        "type": [
          "string",
          "null"
        ]
      },
      "lens": {
        "type": [
          "string",
          "null"
        ]
      },
      "verdict": {
        "description": "banded verdict: primary, confidence, nearCall, distribution"
      },
      "scores": {
        "description": "the five v3.2 dimension scores, 1-5"
      },
      "axes": {
        "description": "strategicDifferentiation (X) and aiFeasibility (Y)"
      },
      "urgency": {},
      "rationale": {
        "type": [
          "string",
          "null"
        ]
      },
      "aiReplacement": {},
      "vendors": {},
      "vendorCount": {
        "type": "number"
      },
      "vendorsWithheld": {
        "type": "number"
      },
      "evidence": {
        "description": "sources and research trail; present only with includeEvidence"
      },
      "evidenceAvailable": {
        "type": "boolean"
      },
      "evidenceHint": {
        "type": "string"
      },
      "guidance": {
        "description": "how to use this result: meaning, action, what not to claim, confidence handling"
      }
    },
    "required": [
      "resolved"
    ],
    "additionalProperties": true,
    "$schema": "http://json-schema.org/draft-07/schema#"
  }
}

Resources 0

No resources advertised.

Prompts 0

No prompts advertised.

Inventory recorded Sep 7, 2026 — 5:05 PM as rev 3. Listing calls only — tools are never invoked.

Canonical contract hash

contract hash — rev 3

sha256:2d4453d5af249683a6ddbe546bb227f6870b1e015809bedc78815ab9cef3c3a9

Computed centrally from the normalized capability inventory (normalizer v1). Scanners submit raw observations; the control plane normalizes and hashes, so every scanner agrees on the same value.

Revisions 3

Rev	Observed	Hash
rev 3	Sep 7, 2026 — 5:05 PM (10 d ago)
rev 2	Aug 9, 2026 — 2:20 PM (39 d ago)
rev 1 baseline	Aug 6, 2026 — 4:55 AM (43 d ago)

Contract changes 2

Non-breaking 5 non-breaking changes. Tool b4_audit now declares an output schema. rev 2 → 3 · 10 d ago

Non-breaking 5 non-breaking changes. Optional parameter org was added at b4_audit.inputSchema. rev 1 → 2 · 39 d ago

Full snapshots are stored only when the canonical hash changes — unchanged observations update freshness without duplicating the contract.

Recent observations latest 30

Observed	Result	Latency
Sep 17, 2026 — 7:35 PM (39 min ago)	Reachable	2099 ms
Sep 17, 2026 — 5:35 PM (3 h ago)	Reachable	124 ms
Sep 17, 2026 — 3:30 PM (5 h ago)	Reachable	143 ms
Sep 17, 2026 — 1:30 PM (7 h ago)	Reachable	447 ms
Sep 17, 2026 — 11:30 AM (9 h ago)	Reachable	223 ms
Sep 17, 2026 — 9:30 AM (11 h ago)	Reachable	119 ms
Sep 17, 2026 — 7:25 AM (13 h ago)	Reachable	198 ms
Sep 17, 2026 — 5:20 AM (15 h ago)	Reachable	135 ms
Sep 17, 2026 — 3:15 AM (17 h ago)	Reachable	138 ms
Sep 17, 2026 — 1:10 AM (19 h ago)	Reachable	125 ms
Sep 16, 2026 — 11:05 PM (21 h ago)	Reachable	265 ms
Sep 16, 2026 — 9:05 PM (23 h ago)	Reachable	424 ms
Sep 16, 2026 — 7:00 PM (1 d ago)	Reachable	159 ms
Sep 16, 2026 — 4:55 PM (1 d ago)	Reachable	138 ms
Sep 16, 2026 — 2:50 PM (1 d ago)	Reachable	439 ms
Sep 16, 2026 — 12:45 PM (1 d ago)	Reachable	124 ms
Sep 16, 2026 — 10:45 AM (1 d ago)	Reachable	751 ms
Sep 16, 2026 — 8:40 AM (1 d ago)	Reachable	183 ms
Sep 16, 2026 — 6:35 AM (2 d ago)	Reachable	138 ms
Sep 16, 2026 — 4:30 AM (2 d ago)	Reachable	141 ms
Sep 16, 2026 — 2:25 AM (2 d ago)	Reachable	288 ms
Sep 16, 2026 — 12:20 AM (2 d ago)	Reachable	156 ms
Sep 15, 2026 — 10:15 PM (2 d ago)	Reachable	799 ms
Sep 15, 2026 — 8:10 PM (2 d ago)	Reachable	130 ms
Sep 15, 2026 — 6:10 PM (2 d ago)	Reachable	202 ms
Sep 15, 2026 — 4:10 PM (2 d ago)	Reachable	192 ms
Sep 15, 2026 — 2:05 PM (2 d ago)	Reachable	240 ms
Sep 15, 2026 — 12:00 PM (2 d ago)	Reachable	130 ms
Sep 15, 2026 — 9:55 AM (2 d ago)	Reachable	167 ms
Sep 15, 2026 — 7:50 AM (3 d ago)	Reachable	156 ms

Latest check — evidence & validation Sep 17, 2026 — 7:35 PM

Reachable MCP Tools Edge Scanner ScheduledHTTP 2002099 msStreamable HTTP2025-06-18

response headers (as observed)

{
  "server": "cloudflare",
  "content-type": "application/json"
}

serverInfo — from initialize

{
  "name": "b4-index",
  "version": "3.3.0",
  "title": null
}

Server name	`b4-index`
Server version	`3.3.0`
Transport	Streamable HTTP
Protocol version	`2025-06-18`
First observed	Aug 6, 2026
Last observed	39 min ago
Last successful	39 min ago
Origin	Registry seed

B4 Index MCP · Ben Roberts AI

Availability — last 30 days

Advertised capabilities

Tools 5

Input schema

Output schema

Input schema

Output schema

Input schema

Output schema

Input schema

Output schema

Input schema

Output schema

Resources 0

Prompts 0

Canonical contract hash

Revisions 3

Contract changes 2

Recent observations latest 30

Publisher supplied

Independently observed

Name	B4 Index MCP
Vendor	Ben Roberts AI
Category	data
Docs	—
Repository	—
Auth expected	API key expected