capture

Captures user content into a brain database and local markdown file with deterministic slugging.

1|Updated May 9, 2026
One-click install
npx skills add https://github.com/weiping/gbrain-cn --skill capture-weiping
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: capture
Source: https://github.com/weiping/gbrain-cn/tree/main/skills/capture
Command: npx skills add https://github.com/weiping/gbrain-cn --skill capture-weiping

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the friction of deciding which ingestion command to use by giving you one reliable CLI entrypoint to store a single idea, snippet, or transcript into your brain.

Core Features & Use Cases

  • Single-command ingestion: Save a thought, article snippet, transcript fragment, or any text using one consistent command.
  • Write-through storage + immediate queryability: Creates a page in the brain database and writes a corresponding markdown file to disk so it’s usable right away.
  • Idempotent, predictable slugs: Re-capturing the same content yields the same inbox/YYYY-MM-DD-<hash8> destination, with dedup handling repeats.
  • Safe trust boundary: This capture path is local-CLI trust (remote: false), while untrusted ingestion goes through a dedicated ingest endpoint.

Quick Start

Run the capture skill to save your text into the brain database and the local inbox file by using: gbrain capture "the thought I want to remember"

Frequently Asked Questions about capture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I save notes and snippets into a knowledge base without picking a specific ingestion workflow?▼

You can capture notes and snippets into a knowledge base using a single CLI entrypoint that writes arbitrary text directly to a brain database. It removes the friction of selecting ingestion commands by storing content immediately as a queryable page.

What is the best way to capture single thoughts and transcripts into a local markdown inbox?▼

Capturing single thoughts or transcripts into a local markdown inbox is done via a single command that accepts inline args, file paths, or stdin. It writes content through to both the brain database and a local markdown file for immediate retrieval.

Does capturing the same note twice create duplicate files in my markdown inbox?▼

Capturing the same note twice does not create duplicate files because the process uses idempotent slugging. Re-capturing identical content yields the same predictable inbox/YYYY-MM-DD-<hash8> destination with dedup handling.

Can I pipe content from standard input to save text into a knowledge database?▼

You can pipe content from standard input to save text into a knowledge database. The capture path accepts inline args, file paths, or stdin, writing the piped data directly to the brain DB and a corresponding local markdown file.

When should I use a local CLI capture command versus a dedicated ingest endpoint for untrusted content?▼

Use the local CLI capture command for trusted content, as it operates with local trust boundaries. Untrusted ingestion should go through a dedicated ingest endpoint, keeping the local capture path safe for your personal brain database.

Why does my saved thought need to be stored in both a database and a markdown file?▼

Saved thoughts are stored in both a database and a markdown file to provide write-through storage and immediate queryability. This ensures the content is usable right away for later synthesis and retrieval from your local inbox.