MCP App Store
by KeenEthics
Platoon Notes icon

Platoon Notes MCP Server for ChatGPT

By Platoon, Inc.

Overview

Platoon Notes gives ChatGPT read-only access to the interview notes and transcripts in your own Platoon workspace: expert interviews, client meetings and other recorded sessions that Platoon has transcribed and summarized. What you can do: - Search across your interviews for a topic and get the matching interviews with deep links into Platoon. - Read the AI notes of an interview (key summary, full note, Q&A script, timeline) and compare several interviews in one request. - Quote a transcript window verbatim around a specific moment. - Browse projects, their tags, documents and questionnaire, and list interviews by date range, type or tag. - Find which interviewees spoke about a topic or match a name or company. - See which meeting-bot sessions are scheduled ahead. Intended users: consultants, researchers and teams who run interviews in Platoon and want to synthesize across them in ChatGPT. Limitations: the connector only reads data the signed-in user already has access to in Platoon. Nothing is created, changed or deleted through it. It does not upload recordings, schedule interviews or share content. A Platoon account with a verified work email is required; sign in with that account when you connect.

How to connect

  1. ChatGPT: Open the Platoon Notes listing in the ChatGPT app directory and choose Connect. ChatGPT listing Platoon Notes exposes 14 tools here.

Tools

date_time

ChatGPT
Current date and time in UTC and KST (Asia/Seoul), with the weekday. Call before turning relative dates ('last week', 'this quarter') into date_from / date_to filters.

fetch

ChatGPT
Fetch one interview by id (from search): its AI-generated notes as text, plus a link and metadata. Transcripts are not included — use get_transcript(interview_id).

find_expert

ChatGPT
Find who to talk to: the experts (interviewees with a profile) who spoke about a topic, or whose name / company / job title matches the query, with their interviews and Links. Inputs: query (topic or name), limit (experts, max 25), optional project_id. Topic matches cover interviews inside your projects (the search index skips personal interviews that have no project; those are still found by name / company / title). Each expert's 'Said' line is something they said, not the interviewer ('From the transcript' when the speaker could not be told); topic-only matches are listed after the name / company / title matches. Interviews without an expert profile are counted separately.

get_account_info

ChatGPT
Who this connector is signed in as: email, organization, role, and the projects it can read. Call this first when results look empty or belong to the wrong workspace.

get_interview

ChatGPT
Get an interview's metadata (status, language, project, speakers) and the list of available notes. Input: interview_id (UUID).

get_notes

ChatGPT
Get the AI-generated notes (e.g. Key Summary, Q&A Script, Full Note) for an interview as hierarchical markdown. Input: interview_id (UUID) or interview_ids (up to 10 UUIDs, rendered back to back — use this to compare interviews in one call); optional note_type to fetch a single note (e.g. key-summary, qna-script, full-note) — prefer it to avoid pulling every note; optional max_length (applies to the whole output).

get_project

ChatGPT
Get one project in depth: interview and member counts, tags (with interview counts), uploaded documents (names only) and the questionnaire template. Input: project_id (UUID from list_projects). Use it to learn a project's structure before searching, or to see which questionnaire items exist.

get_project_notes

ChatGPT
Read one note type across a project's interviews in a single call, newest first. Inputs: project_id (UUID), note_type (default key-summary; e.g. full-note, qna-script, quotes), limit (interviews per call, max 10), offset to page, optional date_from / date_to (YYYY-MM-DD, UTC), max_length for the whole output. Prefer this over many get_notes calls when synthesizing a project.

get_transcript

ChatGPT
Get an interview's transcript as '[bX] Speaker: Text' (or '[bX] Text' with format='text_only'). Input: interview_id (UUID); optional format and max_length. Read a window instead of the whole thing: from_block/to_block (inclusive [bX] indices, as cited in notes) or around_block_id (a block UUID from a search Link's ?blocks=) with context blocks on each side (default 15). Block numbering stays the same in a window, so [bX] citations line up, and a window of at most 200 blocks ends each line with [block_id=<uuid>] for ?blocks= deep links and around_block_id.

list_interviews

ChatGPT
List the signed-in user's interviews, most recent first. Returns interview id, name, project, type, processing status, created date, duration, and a link. Optional filters: project_id (UUID from list_projects), date_from / date_to (YYYY-MM-DD, inclusive, UTC day boundaries, by creation date), interview_type (expert-interview | client-meeting | others), tag (tag name, case-insensitive, matched across your projects). Use search_interviews when looking for a topic rather than browsing.

list_projects

ChatGPT
List the signed-in user's projects (those they are a direct member of), with interview counts. Pass a returned project id to list_interviews' project_id to drill into that project's interviews.

list_upcoming_interviews

ChatGPT
Interviews scheduled ahead: meeting-bot sessions (calendar or manual) in the next N days for interviews you can access, soonest first, with UTC and KST start times. Inputs: days (default 7, max 60), limit (max 50).

search_interviews

ChatGPT
Search across the signed-in user's interviews (notes + transcripts, hybrid keyword/semantic). Use this FIRST to find which interviews discuss a topic; it returns up to limit interviews, each with its id, a deep link, and short matched snippets — not full content. Then call get_notes / get_transcript on the ids you need. Optional filters: project_id (UUID from list_projects), date_from / date_to (YYYY-MM-DD, inclusive, UTC day boundaries, by creation date), interview_type (expert-interview | client-meeting | others), tag (tag name, case-insensitive, matched across your projects).

Capabilities

Search InterviewsRead Interview NotesQuote TranscriptsBrowse Projects

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/platoon-notes"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/platoon-notes)