Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By harshav167
Claude Code plugin for Plaud AI — manage recordings, transcripts, summaries, and audio via CLI commands, MCP tools, prompts, and skill resources
npx claudepluginhub harshav167/plaud-plugin --plugin plaudGet a presigned audio download URL for a recording
Bulk review and export transcripts, summaries, and notes
Trigger Plaud AI transcription and summarization
List Plaud AI recordings with optional filtering
Search recordings by title and memory by content
Guide for bulk ingestion and batch processing of Plaud recordings into the memory search index. Use when the user wants to index multiple recordings, set up auto-ingestion, check job progress, or do batch operations.
Guide for searching across indexed Plaud transcripts using semantic, keyword, graph, and entity extraction. Use when the user asks to search conversation content, find mentions of topics, people, decisions, or action items across recordings.
Guide for finding and browsing Plaud recordings using metadata filters. Use when the user asks to list, find, filter, or browse their recordings by date, source type, title, or transcription status.
Guide for transcribing Plaud recordings using ElevenLabs Scribe v2. Use when the user wants a transcript, asks to transcribe a recording, or needs speaker-diarized text from their recordings. IMPORTANT -- transcription costs API credits.
Guide for using the plaud CLI to manage Plaud AI recordings, transcripts, summaries, and audio files. Use when the user asks about Plaud recordings, transcription, or wants to interact with their Plaud AI data via the command line.
External network access
Connects to servers outside your machine
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Conversation memory for AI assistants — record, transcribe, search meetings and voice memos. 19 skills + 1 agent + 2 hooks.
Claude Code skill pack for Deepgram (24 skills)
Record and transcribe audio to markdown using whisper.cpp with Metal acceleration. Captures mic + system audio.
Session log archiver: saves conversations as Markdown or JSON for easy review
Meeting capture — record mic + system audio, take timestamped notes, transcribe locally, distill into vault-connected notes.
Speaks a short summary of Claude's response using macOS say command
Voice conversations with AI coding assistants using ElevenLabs TTS/STT. HTTP server on port 8765 with local Silero VAD and manual commit mode.
Manage Plaud AI recordings, transcripts, summaries, and memory search directly from Claude Code.
.mcp.json/status -- Daily briefing/ls -- Browse recordings/transcribe <ref> -- Get transcript/search <query> -- Search recordings + memory# Clone into your Claude Code plugins directory
git clone https://github.com/harshav167/plaud-plugin ~/.claude/plugins/plaud
Then edit .mcp.json in the plugin directory to point to your Plaud MCP server:
{
"mcpServers": {
"plaud": {
"type": "http",
"url": "https://your-server.example.com/mcp"
}
}
}
Set the PLAUD_MCP_URL environment variable for the session-check hook:
export PLAUD_MCP_URL="https://your-server.example.com/mcp"
The server uses GitHub OAuth — no manual tokens or env vars needed. Claude handles the OAuth flow automatically when you first connect to the MCP server.
Known limitation: The
agentsfield inplugin.jsonis not supported when installing from a remote marketplace URL (Claude Code validates it as "Invalid input"). Aria's agent file (agents/recording-assistant.md) is present in the repo but is currently excluded fromplugin.jsonto allow installation. To use Aria, copyagents/recording-assistant.mdto~/.claude/agents/manually after installing.
| Command | Description |
|---|---|
/status | Daily briefing -- recent recordings, index health, backlog |
/ls | Browse and filter recordings |
/transcribe | Get transcript (checks cache first) |
/search | Dual-path search (metadata + semantic memory) |
/summary | View AI summary or meeting notes |
/download | Get audio download URL |
/fetch-all | Bulk export recordings |
/generate | Trigger Plaud AI processing |
| Prompt | Purpose |
|---|---|
daily_briefing | Status overview — recent recordings, queue, memory |
analyze_transcript | Deep analysis of a recording |
search_memory | Guided dual-path search |
bulk_summary | Summarize multiple recordings |
| Skill | Scope |
|---|---|
| recording-search | Find and browse recordings |
| memory-search | Semantic search across transcripts |
| transcription | ElevenLabs transcription with cost awareness |
| bulk-operations | Batch ingestion and auto-polling |
| using-plaud-mcp | Quick-start MCP tool reference |
| using-plaud-cli | CLI usage guide |
Skills as Resources: The MCP server exposes all 6 skills as MCP resources (12 total:
SKILL.md+_manifesteach). Any MCP client can discover and read skill documentation without installing the plugin.
All tools are prefixed mcp__plaud__:
| Tool | Purpose | Cost |
|---|---|---|
| find_recordings | Browse/filter recordings | Free |
| get_recording | Recording metadata | Free |
| get_audio_url | Download URL (5min expiry) | Free |
| transcribe | ElevenLabs transcript | Credits* |
| get_content | Summary/notes | Free |
| trigger_processing | Plaud AI processing | Free |
| get_account_info | Account info | Free |
| get_processing_status | Queue status | Free |
| list_languages | Language codes | Free |
| memory_search | Search indexed transcripts | Free |
| memory_ingest | Bulk index recordings | Credits** |
*Cached transcripts return free. **With source="all" only; source="cache" is free.
Session state: Tools accepting file_ref support persistent row-number resolution across calls — call find_recordings once, then use row numbers in subsequent transcribe, get_content, etc. calls within the same session.
Elicitation: transcribe and memory_ingest(source="all") prompt for confirmation before consuming credits. Older MCP clients that don't support elicitation skip the prompt and proceed directly.
This plugin connects to a plaud-cli MCP server. See that repo for server deployment instructions.