Describe Proofable MCP as the trust harness, not a permissions tool

.agents/plugins/marketplace.json

@@ -4,14 +4,14 @@
     "name": "Proofable"
   },
   "metadata": {
-    "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+    "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
     "version": "0.1.7"
   },
   "plugins": [
     {
       "name": "proofable-mcp",
       "source": "./plugins/proofable-mcp",
-      "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run."
+      "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client."
     }
   ]
 }

.claude-plugin/marketplace.json

@@ -4,14 +4,14 @@
     "name": "Proofable"
   },
   "metadata": {
-    "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+    "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
     "version": "0.1.7"
   },
   "plugins": [
     {
       "name": "proofable-mcp",
       "source": "./plugins/proofable-mcp",
-      "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run."
+      "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client."
     }
   ]
 }

.cursor-plugin/marketplace.json

@@ -4,14 +4,14 @@
     "name": "Proofable"
   },
   "metadata": {
-    "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+    "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
     "version": "0.1.7"
   },
   "plugins": [
     {
       "name": "proofable-mcp",
       "source": "./plugins/proofable-mcp",
-      "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run."
+      "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client."
     }
   ]
 }

README.md

@@ -3,9 +3,9 @@
 [![npm](https://img.shields.io/npm/v/%40proofable%2Fmcp?label=%40proofable%2Fmcp&color=98C0EF)](https://www.npmjs.com/package/@proofable/mcp)
 [![License](https://img.shields.io/badge/license-Apache--2.0-blue.svg)](./LICENSE)
 
-Put AI agents on real work without handing over your keys.
+The trust harness for AI, in every MCP client.
 
-Connect once, and every agent you run in Cursor, Claude, Codex, VS Code, ChatGPT or any other MCP client works under a permission you set. Connected apps and payments run through Proofable, are checked before they run, and leave a signed proof anyone can verify.
+Connect Proofable once, and your agents can use your profile, proofs, permissions, and protected tools in Cursor, Claude, Codex, VS Code, ChatGPT, and any other MCP client.
 
 ## Start
 

package.json

@@ -1,7 +1,7 @@
 {
   "name": "@proofable/mcp",
   "version": "0.1.7",
-  "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+  "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
   "type": "module",
   "main": "./index.mjs",
   "exports": {

plugins/proofable-mcp/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "proofable-mcp",
   "displayName": "Proofable",
-  "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+  "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
   "version": "0.1.7",
   "author": {
     "name": "Proofable",

plugins/proofable-mcp/.codex-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "proofable-mcp",
   "version": "0.1.7",
-  "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+  "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
   "author": {
     "name": "Proofable",
     "url": "https://proofable.me"
@@ -29,8 +29,8 @@
   "skills": "./skills/",
   "interface": {
     "displayName": "Proofable",
-    "shortDescription": "Permissions and signed proof for AI agents",
-    "longDescription": "Put AI agents on real work without handing over your keys. Each agent gets a permission for its job, risky steps wait for your approval, and every run leaves a signed proof anyone can verify. One connection for Claude, Cursor, Codex and VS Code.",
+    "shortDescription": "The trust harness for AI",
+    "longDescription": "Bring the same identity, context, limits and proof to every AI client you use. Connected apps and payments are checked before they run and leave a signed proof anyone can verify.",
     "developerName": "Proofable",
     "websiteURL": "https://proofable.me",
     "supportURL": "https://proofable.me/contact",

plugins/proofable-mcp/.cursor-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "proofable-mcp",
   "displayName": "Proofable",
-  "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+  "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
   "version": "0.1.7",
   "author": {
     "name": "Proofable"

plugins/proofable-mcp/README.md

@@ -1,6 +1,6 @@
 # Proofable MCP plugin
 
-Put AI agents on real work without handing over your keys.
+The trust harness for AI, in every MCP client.
 
 This plugin adds the hosted Proofable MCP server and three workflow skills: `proofable-setup`, `proofable-trust-workflow`, and `proofable-integrate`. Install it, finish sign-in in your client, then ask:
 

plugins/proofable-mcp/plugin.json

@@ -2,7 +2,7 @@
   "$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
   "name": "proofable-mcp",
   "version": "0.1.7",
-  "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+  "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
   "author": {
     "name": "Proofable",
     "url": "https://proofable.me"

server.json

@@ -2,7 +2,7 @@
   "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
   "name": "io.github.proofable/mcp",
   "title": "Proofable MCP",
-  "description": "Give AI agents scoped permissions instead of your keys, with approvals and signed proof of each run.",
+  "description": "The trust harness for AI: identity, context, limits and proof for every agent, in any MCP client.",
   "version": "0.1.7",
   "repository": {
     "url": "https://github.com/proofable/mcp",