peepshow/ sinks/ pinecone

Reel #16 Vector DB

peepshow sink / pinecone

PineconeUpsert frames into a Pinecone serverless index.

Serverless. Auto-detects namespace. Embedding of metadata + caption per frame.

drop · process · pinecone

What it does

Push frames into a Pinecone serverless index. Namespaces are configurable so multiple teams/projects can share an index without collisions; IDs map back to the peepshow run + frame index.

When to reach for it

  • Managed, zero-ops vector DB for LLM video pipelines
  • Multi-tenant SaaS — each tenant gets their own Pinecone namespace
  • Pair with OpenAI / Cohere embeddings of the peepshow caption field

Install

npm i -g peepshow

Use it

peepshow ./movie.mp4 --sink pinecone \
  --sink-arg index=peepshow

Make it automatic

Register the sink once — every run fires it afterward. Scope by --when so it only runs for matching videos.

peepshow sinks add pinecone
peepshow sinks add pinecone --when extension=mp4,mov
peepshow sinks add pinecone --when path=/Volumes/Work/

Configuration

  • PEEPSHOW_PINECONE_API_KEY Pinecone API key. required
  • PEEPSHOW_PINECONE_INDEX Index name. required
  • PEEPSHOW_PINECONE_NAMESPACE Namespace. Default `peepshow`.

Use with an LLM agent

Every peepshow sink reads its config from env vars and receives a single JSON payload on stdin. An LLM agent (Claude Code, Cursor, Windsurf, Gemini, Codex) can drive the Pinecone sink automatically when three things are true:

  • the env vars below are exported in the agent's shell (or a project .env it can load),
  • the peepshow CLI is on PATH — install with npm i -g peepshow,
  • a peepshow auto-sink is registered for the run (optional but recommended — makes invocation zero-argument).

1. Set the environment

# Add to ~/.zshrc, ~/.bashrc, or a project .env the agent can load
export PEEPSHOW_PINECONE_API_KEY="..."
export PEEPSHOW_PINECONE_INDEX="..."

2. Register as an auto-sink

peepshow sinks add pinecone
peepshow sinks add pinecone --when extension=mp4,mov

3. Example LLM session

You → drop a .mov into Claude Code.

Claude → auto-invokes /peepshow:slides ./clip.mov. peepshow extracts frames + audio, the Pinecone sink forwards the run to the configured collection. Claude replies with a summary and a link to the created record.

The transcript text is a natural embedding target — index it next to the frames.

Write your own

A sink is any executable that reads the --emit json payload on stdin. Shell, Node, Python, Go — the spec's in docs/PLUGINS.md. Register persistent ones with peepshow sinks add-cmd 'your-command'.