peepshow/ sinks/ chroma

Reel #14 Vector DB

peepshow sink / chroma

ChromaIndex frames into a Chroma vector DB collection.

Text embeddings over the metadata + caption. Frame paths as IDs.

drop · process · chroma

What it does

Push every peepshow run into a Chroma collection as a set of points: one per frame, embedding generated from the caption + container tags, path stored in metadata so you can fetch the JPG back.

When to reach for it

  • RAG over your own video library — query by natural language, get frames back
  • Cross-reference videos by semantic similarity instead of filename
  • Pair a local LLM with a local Chroma for fully-offline video search

Install

npm i -g peepshow

Use it

peepshow ./lecture.mp4 --sink chroma \
  --sink-arg url=http://localhost:8000 \
  --sink-arg collection=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 chroma
peepshow sinks add chroma --when extension=mp4,mov
peepshow sinks add chroma --when path=/Volumes/Work/

Configuration

  • PEEPSHOW_CHROMA_URL Chroma HTTP server URL. required
  • PEEPSHOW_CHROMA_COLLECTION Collection name. 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 Chroma 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_CHROMA_URL="..."

2. Register as an auto-sink

peepshow sinks add chroma
peepshow sinks add chroma --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 Chroma 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'.