@miraigent/free-ai-ops-mcp 0.1.3 → 0.1.5

package/CHANGELOG.md

@@ -1,5 +1,23 @@
 # Changelog
 
+## [0.1.5] - 2026-06-07
+
+### Changed
+
+- Updated LAUNCH_FLOW.md so the package distribution route stays focused on
+  npm/GitHub developers and issue feedback.
+- Removed remaining Agent Memories consumer routing from the npm package files.
+
+## [0.1.4] - 2026-06-07
+
+### Changed
+
+- Repositioned the npm package toward developers, indie hackers, MCP builders,
+  and AI automation builders instead of general creator/audience messaging.
+- Removed Agent Memories consumer links from the npm README so the channel
+  matches npm users.
+- Updated npm metadata to emphasize human-in-the-loop developer tooling.
+
 ## [0.1.3] - 2026-06-07
 
 ### Changed

package/LAUNCH_FLOW.md

@@ -2,8 +2,8 @@
 
 This document defines the first public launch flow for Miraigent Free AI Ops MCP.
 
-The goal is to distribute useful free MCP tools, collect public GitHub issues,
-and guide interested users toward Miraigent and Agent Memories.
+The goal is to distribute useful free MCP tools to developers, collect public
+GitHub issues, and guide implementation-heavy users toward Miraigent.
 
 ## Launch Position
 
@@ -21,7 +21,7 @@ The public promise is:
 - keep CRM notes safer
 - collect feedback in GitHub issues
 - connect deeper implementation needs to Miraigent
-- connect reusable operating lessons to Agent Memories
+- keep the npm package focused on developers who can run npm, npx, or MCP tools
 
 ## First Three MCPs
 
@@ -41,8 +41,8 @@ Why first:
 
 Route:
 
-GitHub free tool -> Issue feedback -> Miraigent free diagnosis for company
-workflow design -> Agent Memories for reusable review lessons.
+GitHub/npm free tool -> Issue feedback -> developer adoption -> Miraigent free
+diagnosis for company workflow design.
 
 ### 2. FAQ Candidate Review MCP
 
@@ -59,8 +59,8 @@ Why second:
 
 Route:
 
-GitHub free tool -> Qiita/Zenn example -> Miraigent resource hub -> free
-diagnosis -> Agent Memories for FAQ decision memory.
+GitHub/npm free tool -> Qiita/Zenn example -> Miraigent resource hub -> free
+diagnosis for FAQ and inquiry-flow design.
 
 ### 3. AI-Safe CRM Notes MCP
 
@@ -77,8 +77,8 @@ Why third:
 
 Route:
 
-GitHub free tool -> note article -> Miraigent free diagnosis -> Agent Memories
-for repeated customer-operation lessons.
+GitHub/npm free tool -> implementation article -> Miraigent free diagnosis for
+real CRM/support workflow design.
 
 ### 4. Prompt Risk Review MCP
 
@@ -96,7 +96,7 @@ Why fourth:
 
 Route:
 
-GitHub free tool -> Issue feedback about missing risk flags -> Miraigent free
+GitHub/npm free tool -> Issue feedback about missing risk flags -> Miraigent free
 diagnosis for data-handling and review-rule design.
 
 ## Second Wave Candidates
@@ -162,19 +162,15 @@ Secondary URL:
 
 https://miraigent.com/resources
 
-### Step 4: Agent Memories Route
+### Step 4: Developer Feedback Route
 
-When a user wants repeated review lessons, decisions, and operating notes to be
-reused, route them to Agent Memories.
+When a developer wants another tool, output field, or review status, route them
+to GitHub issues.
 
 CTA language:
 
-If the same review decisions repeat, turn them into reusable operating memory
-with Agent Memories.
-
-Primary URL:
-
-https://agentmemories.jp/
+If your AI agent or workflow needs another review field, open a public-safe
+issue with the expected input and output.
 
 ## Posting Plan
 
@@ -250,20 +246,19 @@ CTA:
 
 Use Prompt Risk Review MCP and open an issue if a risk flag is missing.
 
-### Post 6: Agent Memories Bridge
+### Post 6: Developer Feedback Bridge
 
 Goal:
 
-Connect repeated decisions to Agent Memories.
+Turn developer use cases into public issues.
 
 Hook:
 
-If the same AI review decision repeats, it should become reusable operating
-memory.
+If your AI agent needs a different review field, make the missing field public.
 
 CTA:
 
-Try the free MCP tools, then use Agent Memories for repeated operating lessons.
+Try the free MCP tools, then open an issue for missing fields or statuses.
 
 ## Suggested First Public Thread
 
@@ -323,10 +318,9 @@ https://miraigent.com/diagnosis
 
 Post 6:
 
-For repeated review lessons and operating decisions, connect the workflow to
-Agent Memories:
+If your AI agent needs another review field, open a public-safe issue:
 
-https://agentmemories.jp/
+https://github.com/Miraigent/miraigent-free-ai-ops-mcp/issues/new/choose
 
 ## Success Signals
 
@@ -338,7 +332,6 @@ Early signals:
 - article views
 - resource hub clicks
 - free diagnosis clicks
-- Agent Memories clicks
 
 Do not over-optimize for stars only. Practical issues from real users are more
 valuable.

package/README.md

@@ -3,8 +3,9 @@
 Free public Model Context Protocol (MCP) server for human-reviewed AI
 operations.
 
-Use it when AI drafts, prompts, FAQ candidates, or CRM notes should pass through
-a visible review gate before they reach customers or shared systems.
+Use it when you are building an AI tool, MCP server, support workflow, or
+internal automation that needs a visible human-review gate before AI output
+reaches customers or shared systems.
 
 ## Try It In 30 Seconds
 
@@ -15,25 +16,25 @@ Run the public MCP server:
 Then send a JSON-RPC tool call such as the copy-ready example in
 examples/human-review-gate/.
 
-If you only want the spreadsheet/checklist version first, download the free
-review kit:
+If you are not a developer and only want the spreadsheet/checklist version
+first, download the free review kit:
 
 https://miraigent.gumroad.com/l/human-review-gate-ai-drafts?utm_source=github&utm_medium=readme&utm_campaign=free-ai-ops-mcp-013
 
 ## Who This Helps
 
-- Support teams that want AI drafts but still need human review.
-- Operators who need prompt risk checks before using customer or business data.
-- Teams building FAQs before they automate inquiry replies.
-- Builders who want small MCP examples before connecting real workflows.
+- Developers adding review gates to AI agents or MCP tools.
+- Indie hackers building support, FAQ, CRM, or prompt-review utilities.
+- AI automation builders who need a safe stop/review/approve step.
+- Teams prototyping human-in-the-loop AI workflows before integrating real data.
+- Operators who can run npm/npx and want a small public MCP example.
 
 ## What To Do Next
 
 - Try one tool with synthetic data.
 - Open an issue if a risk flag, status, or output field is missing.
-- Use the Gumroad kit when you want CSV/checklist files.
-- Use Agent Memories when repeated review decisions should become reusable
-  operating memory.
+- Use the Gumroad kit only if you want CSV/checklist files instead of npm.
+- Adapt the review-gate fields before connecting private systems or real users.
 
 ## Alpha Tools
 
@@ -86,7 +87,7 @@ decision log that match the free Gumroad kit.
 ## Launch Flow
 
 See LAUNCH_FLOW.md for the first public posting plan, issue collection flow, and
-routes toward Miraigent and Agent Memories.
+developer-focused distribution route.
 
 ## Run
 
@@ -139,6 +140,4 @@ This repository must not publish:
 - Resource hub: https://miraigent.com/resources.html
 - Parent template repository: https://github.com/Miraigent/miraigent-ai-ops-templates
 - Free template library: https://miraigent.com/en/free-ai-operations-templates.html
-- Agent Memories MCP catalog: https://github.com/Miraigent/Miraigent-agent-memories-mcp-catalog?utm_source=github&utm_medium=readme&utm_campaign=free-ai-ops-mcp-013
-- Agent Memories: https://agentmemories.jp/?utm_source=github&utm_medium=readme&utm_campaign=free-ai-ops-mcp-013
 - Free Gumroad review kit: https://miraigent.gumroad.com/l/human-review-gate-ai-drafts?utm_source=github&utm_medium=readme&utm_campaign=free-ai-ops-mcp-013

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

@@ -1,6 +1,6 @@
 #!/usr/bin/env node
 
-const serverInfo = { name: 'miraigent-free-ai-ops-mcp', version: '0.1.3' };
+const serverInfo = { name: 'miraigent-free-ai-ops-mcp', version: '0.1.5' };
 
 const tools = [
   {

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@miraigent/free-ai-ops-mcp",
-  "version": "0.1.3",
-  "description": "Free Model Context Protocol MCP server for human review gates, prompt risk checks, FAQ review, and AI support operations.",
+  "version": "0.1.5",
+  "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",
   "private": false,
@@ -39,10 +39,11 @@
     "ai-safety",
     "crm",
     "faq",
-    "agent-memory",
     "ai-agent",
     "workflow-automation",
-    "review-gate"
+    "review-gate",
+    "human-in-the-loop",
+    "developer-tools"
   ],
   "repository": {
     "type": "git",