@plaud-ai/mcp 0.1.19 → 0.1.21
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.
- package/package.json +3 -2
- package/plugin.json +1 -1
- package/skills/plaud/SKILL.md +34 -0
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@plaud-ai/mcp",
|
|
3
|
-
"version": "0.1.
|
|
3
|
+
"version": "0.1.21",
|
|
4
4
|
"type": "module",
|
|
5
5
|
"main": "dist/index.js",
|
|
6
6
|
"bin": {
|
|
@@ -9,7 +9,8 @@
|
|
|
9
9
|
"files": [
|
|
10
10
|
"dist",
|
|
11
11
|
"plugin.json",
|
|
12
|
-
".mcp.json"
|
|
12
|
+
".mcp.json",
|
|
13
|
+
"skills"
|
|
13
14
|
],
|
|
14
15
|
"scripts": {
|
|
15
16
|
"version:show": "node -p \"require('./package.json').version\"",
|
package/plugin.json
CHANGED
|
@@ -0,0 +1,34 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: plaud
|
|
3
|
+
description: Access and manage Plaud recordings, transcriptions, and meeting notes. Use when the user mentions Plaud, wants to find recordings, review transcripts, check meeting summaries, or manage their audio files.
|
|
4
|
+
---
|
|
5
|
+
|
|
6
|
+
You have access to the user's Plaud recordings via MCP tools. Use them to help the user find, review, and analyze their recordings.
|
|
7
|
+
|
|
8
|
+
## Available tools
|
|
9
|
+
|
|
10
|
+
| Tool | When to use |
|
|
11
|
+
|---|---|
|
|
12
|
+
| `login` | User is not authenticated, or auth has expired |
|
|
13
|
+
| `logout` | User wants to sign out or disconnect |
|
|
14
|
+
| `get_current_user` | User asks who is logged in, or to verify account |
|
|
15
|
+
| `list_files` | User wants to browse, search, or find recordings |
|
|
16
|
+
| `get_file` | User wants details, transcript, or summary of a specific recording |
|
|
17
|
+
|
|
18
|
+
## Typical workflows
|
|
19
|
+
|
|
20
|
+
**First-time use:** Call `login` first. If any tool returns an auth error, call `login` before retrying.
|
|
21
|
+
|
|
22
|
+
**Find a recording:** Call `list_files` to browse. Use pagination (`page`, `page_size`) if the user wants more results.
|
|
23
|
+
|
|
24
|
+
**Review a recording:** Call `get_file` with the file ID. The response includes:
|
|
25
|
+
- `source_list`: timestamped transcript with speaker labels
|
|
26
|
+
- `note_list`: AI-generated summaries in Markdown
|
|
27
|
+
- `presigned_url`: temporary download link (valid 24 hours)
|
|
28
|
+
|
|
29
|
+
## Guidelines
|
|
30
|
+
|
|
31
|
+
- Always check if the user is logged in before calling data tools. If you get an auth error, call `login` first.
|
|
32
|
+
- When listing files, present results clearly with name, date, and duration.
|
|
33
|
+
- When showing transcripts, format them readably with timestamps and speaker labels.
|
|
34
|
+
- When showing notes, render the Markdown content directly.
|