md-feedback 0.8.0 → 0.9.3

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 (4) hide show
  1. package/README.md +76 -72
  2. package/bin/md-feedback.cjs +19 -19
  3. package/package.json +9 -5
  4. package/dist/mcp-server.js +0 -22435
package/README.md CHANGED
@@ -1,72 +1,76 @@
1
- # md-feedback
2
-
3
- > MCP server for markdown annotation review — AI agents read your annotations directly.
4
-
5
- [![npm](https://img.shields.io/npm/v/md-feedback)](https://www.npmjs.com/package/md-feedback)
6
- [![License: SUL-1.0](https://img.shields.io/badge/License-SUL--1.0-blue.svg)](https://github.com/yeominux/md-feedback/blob/dev/LICENSE)
7
-
8
- ## What is this?
9
-
10
- md-feedback is an [MCP](https://modelcontextprotocol.io/) server that lets AI agents (Claude Code, Cursor, Copilot, etc.) read your markdown review annotations, mark memos done, evaluate quality gates, and generate session handoffs — all automatically.
11
-
12
- **This is the MCP server component.** For the VS Code extension, see [MD Feedback on VS Code Marketplace](https://marketplace.visualstudio.com/items?itemName=yeominux.md-feedback-vscode).
13
-
14
- ## Quick Start
15
-
16
- Add to your MCP client config (Claude Code, Cursor, etc.):
17
-
18
- ```json
19
- {
20
- "mcpServers": {
21
- "md-feedback": {
22
- "command": "npx",
23
- "args": ["-y", "md-feedback"]
24
- }
25
- }
26
- }
27
- ```
28
-
29
- That's it. No install, no setup — `npx` handles everything.
30
-
31
- ## 12 MCP Tools
32
-
33
- | Tool | Description |
34
- |------|-------------|
35
- | `get_document_structure` | Full review state: memos, gates, cursor, sections, summary |
36
- | `list_annotations` | All annotations with type/status/owner/color |
37
- | `get_review_status` | Annotation counts and session status |
38
- | `update_memo_status` | Mark a memo as open/answered/done/wontfix |
39
- | `update_cursor` | Set plan cursor position (task ID, step, next action) |
40
- | `evaluate_gates` | Check if merge/release/implement conditions are met |
41
- | `export_review` | Export for a specific AI tool format |
42
- | `create_checkpoint` | Save review progress |
43
- | `get_checkpoints` | List all checkpoints |
44
- | `generate_handoff` | Generate structured handoff document |
45
- | `pickup_handoff` | Parse existing handoff for session resumption |
46
- | `create_annotation` | Create annotation programmatically |
47
-
48
- ## How It Works
49
-
50
- 1. You annotate a markdown plan in the [VS Code extension](https://marketplace.visualstudio.com/items?itemName=yeominux.md-feedback-vscode)
51
- 2. AI agent reads annotations via MCP — no export needed
52
- 3. Agent implements fixes, marks memos done
53
- 4. Gates auto-evaluate — agent knows when it's safe to merge
54
- 5. Agent generates handoff — next session picks up where you left off
55
-
56
- ## CLI
57
-
58
- ```bash
59
- md-feedback --help # Show help
60
- md-feedback --version # Print version
61
- md-feedback # Start MCP server (stdio)
62
- ```
63
-
64
- ## Links
65
-
66
- - [VS Code Extension](https://marketplace.visualstudio.com/items?itemName=yeominux.md-feedback-vscode)
67
- - [GitHub](https://github.com/yeominux/md-feedback)
68
- - [Documentation](https://github.com/yeominux/md-feedback#readme)
69
-
70
- ## License
71
-
72
- [SUL-1.0](https://github.com/yeominux/md-feedback/blob/dev/LICENSE) — Free for personal and non-commercial use.
1
+ # md-feedback
2
+
3
+ > MCP server for markdown annotation review — AI agents read your annotations directly.
4
+
5
+ [![npm](https://img.shields.io/npm/v/md-feedback)](https://www.npmjs.com/package/md-feedback)
6
+ [![License: SUL-1.0](https://img.shields.io/badge/License-SUL--1.0-blue.svg)](https://github.com/yeominux/md-feedback/blob/main/LICENSE)
7
+
8
+ ## What is this?
9
+
10
+ md-feedback is an [MCP](https://modelcontextprotocol.io/) server that lets AI agents (Claude Code, Cursor, Copilot, etc.) read your markdown review annotations, mark memos done, evaluate quality gates, and generate session handoffs — all automatically.
11
+
12
+ **This is the MCP server component.** For the VS Code extension, see [MD Feedback on VS Code Marketplace](https://marketplace.visualstudio.com/items?itemName=yeominux.md-feedback-vscode).
13
+
14
+ ## Quick Start
15
+
16
+ Add to your MCP client config (Claude Code, Cursor, etc.):
17
+
18
+ ```json
19
+ {
20
+ "mcpServers": {
21
+ "md-feedback": {
22
+ "command": "npx",
23
+ "args": ["-y", "md-feedback"]
24
+ }
25
+ }
26
+ }
27
+ ```
28
+
29
+ That's it. No install, no setup — `npx` handles everything.
30
+
31
+ ## 12 MCP Tools
32
+
33
+ | Tool | Description |
34
+ |------|-------------|
35
+ | `get_document_structure` | Full review state: memos, gates, cursor, sections, summary |
36
+ | `list_annotations` | All annotations with type/status/owner/color |
37
+ | `get_review_status` | Annotation counts and session status |
38
+ | `update_memo_status` | Mark a memo as open/answered/done/wontfix |
39
+ | `update_cursor` | Set plan cursor position (task ID, step, next action) |
40
+ | `evaluate_gates` | Check if merge/release/implement conditions are met |
41
+ | `export_review` | Export for a specific AI tool format |
42
+ | `create_checkpoint` | Save review progress |
43
+ | `get_checkpoints` | List all checkpoints |
44
+ | `generate_handoff` | Generate structured handoff document |
45
+ | `pickup_handoff` | Parse existing handoff for session resumption |
46
+ | `create_annotation` | Create annotation programmatically |
47
+
48
+ ## How It Works
49
+
50
+ 1. You annotate a markdown plan in the [VS Code extension](https://marketplace.visualstudio.com/items?itemName=yeominux.md-feedback-vscode)
51
+ 2. AI agent reads annotations via MCP — no export needed
52
+ 3. Agent implements fixes, marks memos done
53
+ 4. Gates auto-evaluate — agent knows when it's safe to merge
54
+ 5. Agent generates handoff — next session picks up where you left off
55
+
56
+ ## CLI
57
+
58
+ ```bash
59
+ md-feedback --help # Show help
60
+ md-feedback --version # Print version
61
+ md-feedback # Start MCP server (stdio)
62
+ ```
63
+
64
+ ## Who Is This For?
65
+
66
+ Developers and team leads who use AI coding agents (Claude Code, Cursor, Copilot) and want to review AI-generated plans before implementation — not after.
67
+
68
+ ## Links
69
+
70
+ - [VS Code Extension](https://marketplace.visualstudio.com/items?itemName=yeominux.md-feedback-vscode)
71
+ - [GitHub](https://github.com/yeominux/md-feedback)
72
+ - [Documentation](https://github.com/yeominux/md-feedback#readme)
73
+
74
+ ## License
75
+
76
+ [SUL-1.0](https://github.com/yeominux/md-feedback/blob/main/LICENSE) — Free for personal and non-commercial use.
package/bin/md-feedback.cjs CHANGED
@@ -1,19 +1,19 @@
1
- #!/usr/bin/env node
2
-
3
- const arg = process.argv[2]
4
- if (arg === '--help' || arg === '-h') {
5
- console.log('md-feedback - MCP server for markdown annotation review\n')
6
- console.log('Usage: md-feedback Start MCP server (stdio transport)')
7
- console.log(' md-feedback --version Print version')
8
- console.log(' md-feedback --help Show this help\n')
9
- console.log('Configure in your AI tool\'s MCP settings:')
10
- console.log(' { "command": "npx", "args": ["-y", "md-feedback"] }')
11
- process.exit(0)
12
- }
13
- if (arg === '--version' || arg === '-v') {
14
- const pkg = require('../package.json')
15
- console.log(pkg.version)
16
- process.exit(0)
17
- }
18
-
19
- require('../dist/mcp-server.js')
1
+ #!/usr/bin/env node
2
+
3
+ const arg = process.argv[2]
4
+ if (arg === '--help' || arg === '-h') {
5
+ console.log('md-feedback - MCP server for markdown annotation review\n')
6
+ console.log('Usage: md-feedback Start MCP server (stdio transport)')
7
+ console.log(' md-feedback --version Print version')
8
+ console.log(' md-feedback --help Show this help\n')
9
+ console.log('Configure in your AI tool\'s MCP settings:')
10
+ console.log(' { "command": "npx", "args": ["-y", "md-feedback"] }')
11
+ process.exit(0)
12
+ }
13
+ if (arg === '--version' || arg === '-v') {
14
+ const pkg = require('../package.json')
15
+ console.log(pkg.version)
16
+ process.exit(0)
17
+ }
18
+
19
+ require('../dist/mcp-server.js')
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "md-feedback",
3
- "version": "0.8.0",
4
- "description": "MCP server for MD Feedback — AI agents read your markdown annotations, mark memos done, evaluate gates, and generate handoffs. No export step needed.",
3
+ "version": "0.9.3",
4
+ "description": "MCP server for MD Feedback: AI agents read markdown review annotations, update memo status, evaluate gates, and generate handoffs.",
5
5
  "license": "SUL-1.0",
6
6
  "author": "Yeomin Seon",
7
7
  "type": "commonjs",
@@ -31,9 +31,11 @@
31
31
  "type": "git",
32
32
  "url": "https://github.com/yeominux/md-feedback"
33
33
  },
34
- "homepage": "https://github.com/yeominux/md-feedback#mcp-server--agent-memory",
34
+ "homepage": "https://github.com/yeominux/md-feedback#mcp-server",
35
35
  "keywords": [
36
36
  "mcp",
37
+ "mcp-server",
38
+ "model-context-protocol",
37
39
  "markdown",
38
40
  "feedback",
39
41
  "ai",
@@ -43,14 +45,16 @@
43
45
  "ai-agent",
44
46
  "coding-workflow",
45
47
  "handoff",
48
+ "session-handoff",
46
49
  "structured-feedback",
47
50
  "checkpoint",
48
51
  "ai-context",
49
52
  "ai-coding",
50
- "mcp-server",
51
53
  "claude-code",
52
54
  "cursor-ai",
55
+ "vibe-coding",
56
+ "context-engineering",
53
57
  "gates",
54
- "plan-cursor"
58
+ "plan-review-tool"
55
59
  ]
56
60
  }