@miraigent/free-ai-ops-mcp 0.1.12 → 0.1.14

package/CHANGELOG.md

@@ -1,5 +1,22 @@
 # Changelog
 
+## [0.1.14] - 2026-06-18
+
+### Changed
+
+- Added a one-command npx smoke-test path to README so npm and GitHub visitors
+  can verify the public MCP JSON-RPC session before editing desktop client
+  settings.
+- Added npm discovery keywords for JSON-RPC and MCP JSON-RPC searches.
+
+## [0.1.13] - 2026-06-17
+
+### Changed
+
+- Added a README feedback-result example so developers who try the npx package
+  can open safer, more actionable GitHub issues with synthetic input shapes and
+  returned status fields.
+
 ## [0.1.12] - 2026-06-16
 
 ### Changed

package/README.md

@@ -19,6 +19,24 @@ examples/human-review-gate/.
 For a complete public-safe MCP session that covers initialize, tools/list, and
 one tool call, see examples/mcp-json-rpc-session/.
 
+## One-Command Public Smoke Test
+
+Use this when you want to confirm the npm package works before editing a desktop
+MCP client config:
+
+    npx -y @miraigent/free-ai-ops-mcp < examples/mcp-json-rpc-session/sample-session.jsonl
+
+The response should include:
+
+- serverInfo.name: miraigent-free-ai-ops-mcp
+- tools/list with four tools
+- human_review_gate returning `stop` for the synthetic refund/complaint/personal
+  data example
+
+This is the fastest public proof path for npm and GitHub users. If the command
+starts but the returned status is confusing, open a Tried It feedback issue with
+the command, synthetic input shape, and returned status only.
+
 If the result is unclear, open a Tried It feedback issue with synthetic input
 only. A useful issue can be as small as:
 
@@ -75,6 +93,27 @@ you ran, the public-safe input shape, and the returned status field. Do not
 paste private customer records, secrets, internal policy text, or paid product
 files.
 
+## Copy A Useful Feedback Result
+
+When you open a Tried It feedback issue, the most useful report is the smallest
+public-safe result shape that shows where the review gate helped or felt unclear.
+
+Good public-safe issue body:
+
+    Command: npx @miraigent/free-ai-ops-mcp
+    Tool: human_review_gate
+    Input shape:
+      draftType: customer support reply
+      audience: customer
+      riskFlags: refund; complaint; personal data
+      sendMode: manual
+    Returned status: stop
+    Confusing part: I expected review_required, but stop may be correct because refund and personal data were both present.
+    Requested improvement: Add one sentence explaining why stop is stronger than review_required.
+
+Keep the report synthetic. Replace names, emails, customer text, internal policy
+text, and private CRM notes with short labels before posting.
+
 ## Who This Helps
 
 - Developers adding review gates to AI agents or MCP tools.

package/mcp/free-ai-ops-server.mjs

@@ -1,6 +1,6 @@
 #!/usr/bin/env node
 
-const serverInfo = { name: 'miraigent-free-ai-ops-mcp', version: '0.1.12' };
+const serverInfo = { name: 'miraigent-free-ai-ops-mcp', version: '0.1.14' };
 
 const tools = [
   {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@miraigent/free-ai-ops-mcp",
-  "version": "0.1.12",
+  "version": "0.1.14",
   "description": "Free MCP server for developers adding human review gates, prompt risk checks, FAQ review, and CRM note safety to AI tools.",
   "homepage": "https://github.com/Miraigent/miraigent-free-ai-ops-mcp",
   "type": "module",
@@ -50,7 +50,9 @@
     "cursor-mcp",
     "mcp-tools",
     "ai-agent-tools",
-    "npx"
+    "npx",
+    "json-rpc",
+    "mcp-json-rpc"
   ],
   "repository": {
     "type": "git",