@postcept/mcp - npm Package Compare versions

Comparing version

+2

-2

dist/index.js

		#!/usr/bin/env node
		// Postcept MCP server. Exposes Proof-of-Completion to AI agents as MCP tools.
		// Postcept MCP server. Exposes outcome verification to AI agents as MCP tools.
		// After an agent performs a high-risk action (a refund, a cancellation, a ticket
		@@ -66,3 +66,3 @@ // resolution), it calls the matching verify_* tool to prove the action actually
		const server = new McpServer({ name: "postcept", version: VERSION }, {
		instructions: "Postcept provides Proof-of-Completion for AI agents. After your agent performs a " +
		instructions: "Postcept provides outcome verification for AI agents. After your agent performs a " +
		"high-risk action (a refund, a subscription cancellation, or a support-ticket " +
		@@ -69,0 +69,0 @@ "resolution), call the matching verify_* tool to confirm the action actually completed " +

+3

-3

package.json

		{
		"name": "@postcept/mcp",
		"version": "0.3.0",
		"description": "Model Context Protocol server for Postcept, Proof-of-Completion for AI agents. Lets an agent verify its own high-risk actions (refunds, cancellations, ticket resolutions) against the system of record and receive a signed completion receipt.",
		"version": "0.3.1",
		"description": "Model Context Protocol server for Postcept, outcome verification for AI agents. Lets an agent verify its own high-risk actions (refunds, cancellations, ticket resolutions) against the system of record and receive a signed completion receipt.",
		"license": "MIT",
		@@ -22,3 +22,3 @@ "author": "Postcept",
		},
		"homepage": "https://github.com/Postcept/mcp#readme",
		"homepage": "https://postcept.com",
		"bugs": "https://github.com/Postcept/mcp/issues",
		@@ -25,0 +25,0 @@ "publishConfig": {

+15

-1

README.md

		# @postcept/mcp

		A [Model Context Protocol](https://modelcontextprotocol.io) server that exposes
		Postcept: Proof-of-Completion for AI agents as MCP tools.
		Postcept, outcome verification for AI agents, as MCP tools.

		@@ -56,2 +56,16 @@ After an agent performs a high-risk action (a refund, a subscription

		## Advisory, by design

		MCP puts the verification call in the agent's hands, which makes it the right
		integration for development, demos, and workflows where the agent's judgment is
		already trusted. It is advisory: an agent can skip the call, pass the wrong
		identifiers, or ignore the answer.

		For consequential actions, enforce the decision outside the agent. The workflow
		or orchestrator calls the Postcept API (or the SDK's `guard()`) itself and
		branches on `safe_to_claim_complete` before anything customer-facing happens.
		The agent then receives the allowed outcome instead of deciding it. Both
		patterns use the same API and the same receipts, what changes is who owns the
		branch.

		## Development
		@@ -58,0 +72,0 @@

Improved metrics