brain-ingest-gate

Validates named-entity resolution and semantic dedup before writing pages into a gbrain knowledge base.

29.4k|4.4k|Updated Apr 5, 2026
One-click install
npx skills add https://github.com/garrytan/gbrain --skill brain-ingest-gate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brain-ingest-gate
Source: https://github.com/garrytan/gbrain/tree/main/plugin/skills/brain-ingest-gate
Command: npx skills add https://github.com/garrytan/gbrain --skill brain-ingest-gate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Content entering a gbrain knowledge base often creates duplicate pages: raw file copies, reworded insights, and named entities stored under generic phrases all fragment the brain. This Skill enforces a pre-write quality gate so every new page is resolved against existing entities and checked for semantic duplicates before it lands.

Core Features & Use Cases

  • Named-Entity Resolution Gate: Resolves people, companies, projects, and concepts registry-first via gbrain entity, expands through alias lists, and reads candidate pages before concluding anything is new.
  • Semantic Dedup Gate: Extracts the core claim, searches the brain, opens and reads the top hit, and assigns a clear-dup / plausible-dup / clear band that decides write, link, or skip.
  • Delegation and Verification: Routes approved writes to shipped enrichment skills (ingest, enrich, signal-detector, concept-synthesis) and verifies results with gbrain check-backlinks check.
  • Use Case: When asked to migrate a folder of project notes into the brain, the gate checks each file for existing entity pages and duplicate insights, links duplicates to canonical pages, and only writes genuinely new content with proper aliases and backlinks.

Quick Start

Ask the agent to move these notes into the brain but check for duplicates and existing entity pages before writing anything.

Frequently Asked Questions about brain-ingest-gate

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

FAQPage Schema
How do I check for duplicates before writing to a gbrain knowledge base?

Extract the core claim of the new content, run gbrain search with a limit of 5, then open and read the top hit with gbrain get. Assign a band of clear-dup, plausible-dup, or clear based on reading, and link instead of writing when the insight already exists.

How do I migrate files into a gbrain brain without creating duplicates?

Run each file through the ingest gate before any gbrain sync or import: resolve named entities with gbrain entity first, then run the dedup decision tree. For batches, gate 3-5 items and inspect the decisions before processing the rest.

Why does gbrain import not catch duplicate content?

gbrain import and sync only skip pages with matching frontmatter IDs, so identical content under a different slug indexes twice. gbrain capture dedupes only identical bytes within 24 hours, not reworded insights, so semantic dedup requires this gate.

Can I trust vector similarity scores to detect duplicate named entities?

No. Pages stored under a chosen name do not embed close to generic descriptive phrases, so a low score does not prove absence. Resolve registry-first with gbrain entity, expand through aliases, and read any plausible candidate page before concluding.

When should dedup be skipped during brain ingestion?

Skip semantic dedup for operational state files, meeting transcripts, timeline entries on existing pages, and media files. These are unique by definition or deduped by filename and hash rather than semantic similarity.