Memory

streaming-audio-memory

Ingest live edge-device audio (car, doorbell, wearable, kiosk) into per-device streaming memory and recall what a device heard. Use when an agent streams audio chunks from devices, monitors ambient audio, or answers questions about a device's audio history. Requires a BlueColumn API key (bc_live_*).

What it does

The skill document

Streaming Audio Memory — BlueColumn Skill

Stream audio chunks from any edge device into persistent, per-device memory. Each chunk is transcribed (Whisper large-v3), mined for entities and intent, and indexed under _audio_streaming_ for instant recall.

Setup

Read TOOLS.md or the platform secret store for the BlueColumn API key (bc_live_*). Base URL: https://xkjkwqbfvkswwdmbtndo.supabase.co/functions/v1

Ingest a chunk

curl -X POST .../streaming-audio \
  -H "Authorization: Bearer " \
  -H "Content-Type: application/json" \
  -d '{"deviceId": "dashcam-01", "audio": "", "format": "wav",
       "sampleRate": 16000, "timestamp": 1756000000000,
       "idempotencyKey": "chunk_dashcam-01_1756000000000_a1b2c3d4e5f6"}'

Idempotency key format: chunk___<6-12 hex>. Same key + same body returns the cached response; same key + different body = 409. Retries never double-process.

Recall what a device heard

curl -X POST .../streaming-audio/query \
  -H "Authorization: Bearer " \
  -H "Content-Type: application/json" \
  -d '{"deviceId": "dashcam-01", "query": "what happened near the warehouse?"}'

Add ?stream=1 (or Accept: text/event-stream) for SSE: sources → delta → done events.

Pipeline (per chunk)

Transcription — Groq Whisper large-v3
Entity extraction — people / locations / events / actions
Intent classification — query | command | notification | warning | conversation
Persist — indexed segment in the device namespace
Session summary — auto-summarized after 30s silence or 5 min of stream

Workflow

Buffer device audio client-side in 5–30s chunks (WAV/Opus/PCM/MP3)
POST each chunk with a fresh idempotency key
On user/environment questions, hit /streaming-audio/query for that device
Surface session_summary memories for long-horizon context

Related skills

bluecolumn-memory

Give AI agents persistent semantic memory using the BlueColumn API (bluecolumn.ai). Use when asked to remember, store, recall, or search memory using BlueColumn; when ingesting notes, conversations, documents, or audio into BlueColumn; when querying what an agent has previously stored; or when wiring up BlueColumn memory endpoints (/agent-remember, /agent-recall, /agent-note) in any workflow. Also use when the user mentions their BlueColumn API key (bc_live_*) and wants to store or retrieve information.

14 installs

voice-call-memory

Memory for voice agents and voice-first workflows — every call transcribed, summarized, and searchable by caller, topic, and outcome. Use when a voice agent must remember callers across conversations. Requires a BlueColumn API key (bc_live_*).

voice-memory

Your voice agent remembers forever. Give voice agents automatic memory of every conversation using BlueColumn. Use when a voice/phone agent must remember callers, meetings, coaching sessions, sales calls, or journal entries; when wiring voice transcripts into BlueColumn; or when a caller needs continuity across calls. Requires a BlueColumn API key (bc_live_*).

1 stars

voice-agent-memory

Full-stack voice agent with BlueColumn persistent memory. Make and receive phone calls with real-time transcription, automatic vector memory storage, and cross-call recall. Every call remembers who called, what they said, and what happened last time.

meeting-memory

Give AI agents Every meeting becomes searchable. using BlueColumn persistent memory. Use when an agent takes meeting notes and needs recall; when the user wants to store, recall, or search meeting memory context. Requires a BlueColumn API key (bc_live_*).

13 installs

BlueColumn ClawHub Integration

Integrate ClawHub with BlueColumn AI for audio ingest, memory, and semantic retrieval. Use when wiring ClawHub to push calls, music, podcasts, and voice notes through BlueColumn's /v1/audio/ingest endpoint; when setting up audio pipeline from ClawHub to BlueColumn; or when configuring real-time audio streaming for live conversations with mid-call memory.

1 installs