aibvf-mcp 0.10.0 → 0.10.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (3) hide show
  1. package/README.md +11 -2
  2. package/dist/server.js +1 -1
  3. package/package.json +3 -3
package/README.md CHANGED
@@ -1,10 +1,18 @@
1
1
  # aibvf-mcp
2
2
 
3
- MCP server exposing AI BVF v1.0 to any Claude agent. Pre-flight-check your AI initiatives before you deploy them.
3
+ MCP server exposing AI BVF v1.0 to any Claude agent, nine deterministic tools that pre-flight-check AI initiatives before the budget is committed: score honestly from whatever is known (a fully estimated pass can never return Accelerate), return the change plan when the verdict is Fix, and measure organisational readiness from process data instead of self-report.
4
4
 
5
5
  > **Source:** [github.com/Bahamas1717/ai-bvf](https://github.com/Bahamas1717/ai-bvf) · ⭐ star if this helped · [Issues](https://github.com/Bahamas1717/ai-bvf/issues) · Built by [Craig Horton Advisory](https://craighortonadvisory.com)
6
6
 
7
- ## Install and run
7
+ ## No install: use it on claude.ai
8
+
9
+ Settings, then Connectors, then Add custom connector, and paste the hosted endpoint. Works on web and mobile, all nine tools, same deterministic engine:
10
+
11
+ ```
12
+ https://mcp.aibvf.com/api/mcp
13
+ ```
14
+
15
+ ## Install and run (stdio)
8
16
 
9
17
  ```bash
10
18
  npx aibvf-mcp
@@ -83,6 +91,7 @@ Claude will call `score_initiative` and return the classification, euro range, a
83
91
  | `get_benchmark` | Return the published benchmark base-rate and industry multiplier for a function + industry. Use when the caller wants the raw rates without an initiative-level verdict. |
84
92
  | `list_taxonomy` | List the valid industries, functions, AI tiers, and readiness levels. |
85
93
  | `diagnose_process` | AI BVF Advisor Brain: diagnose one business process from observed signals (volume, labour, cycle time, handoffs, rework, automation, spend) and return heaviness, the recommended intervention (Automate / Consolidate & re-sequence / Quality controls / Eliminate), the modelled net EUR saving, the efficiency gain, an Accelerate/Fix/Stop verdict, and a decision confidence governed by how much was actually measured. |
94
+ | `infer_readiness` | Measures organisational readiness from process signals (hand-offs, rework, touch ratio, automation, cycle time vs function medians) instead of accepting self-report. Returns the classification the data supports, per-signal reasoning, and a confidence set by coverage and agreement. When the measured answer is lower than the claimed one, that gap is itself a change-readiness finding. |
86
95
 
87
96
  ## Spec
88
97
 
package/dist/server.js CHANGED
@@ -120,7 +120,7 @@ export function logCall(tool_name, meta = {}) {
120
120
  });
121
121
  }
122
122
  /** Single source of truth for the server version, shared by both transports. */
123
- export const VERSION = '0.10.0';
123
+ export const VERSION = '0.10.1';
124
124
  const scoreInputSchema = {
125
125
  type: 'object',
126
126
  required: ['industry', 'revenue_eur', 'function', 'ai_tier', 'readiness'],
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "aibvf-mcp",
3
3
  "mcpName": "io.github.Bahamas1717/aibvf-mcp",
4
- "version": "0.10.0",
4
+ "version": "0.10.1",
5
5
  "description": "AI BVF: score AI portfolios Stop/Fix/Accelerate with decision confidence and pace-layer drag.",
6
6
  "keywords": [
7
7
  "mcp",
@@ -42,8 +42,8 @@
42
42
  "@modelcontextprotocol/sdk": "^1.0.0"
43
43
  },
44
44
  "devDependencies": {
45
- "typescript": "^5.4.0",
46
- "@types/node": "^20.0.0"
45
+ "typescript": "^6.0.3",
46
+ "@types/node": "^26.1.0"
47
47
  },
48
48
  "exports": {
49
49
  "./server": {