brain-ops

Detects entities in inbound signals and queries the brain before external research.

Updated Apr 23, 2026
One-click install
npx skills add https://github.com/ngochuy13/intern-dev --skill brain-ops-ngochuy13
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brain-ops
Source: https://github.com/ngochuy13/intern-dev/tree/main/skills/brain-ops
Command: npx skills add https://github.com/ngochuy13/intern-dev --skill brain-ops-ngochuy13

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents the agent from researching or answering from scratch by enforcing a brain-first read/enrich/write loop for every entity-related interaction.

Core Features & Use Cases

  • Brain-first lookup: Searches and queries the brain before any external research on people, companies, or topics.
  • Read → enrich → write loop: Detects entities in inbound signals, loads relevant pages, writes back new information with attribution, and updates timeline context.
  • Mandatory back-linking: Ensures every mention of a person or company is linked back from the entity’s page to the mentioning page to avoid broken knowledge graph trails.
  • Ambient enrichment: Runs continuously so signals can spawn background enrichment without interrupting the user.

Quick Start

Use brain-ops to look up a person or company in the brain first, then enrich and write any new facts back with proper source attribution and back-links.

Frequently Asked Questions about brain-ops

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

FAQPage Schema
How do I ground agent answers in a knowledge graph for entity-related questions?

Ground agent answers by enforcing a brain-first read, enrich, and write loop that queries an internal context store before any external research. This workflow ensures every entity-centric response references existing knowledge rather than generating answers from scratch.

What is ambient enrichment and how does it work for entity resolution?

Ambient enrichment runs continuously in the background so inbound signals can spawn automatic entity detection, page loading, and graph reconciliation without interrupting the user. It enriches people, company, or topic data seamlessly while maintaining source attribution.

How do I prevent broken knowledge graph trails when writing back entity information?

Prevent broken knowledge graph trails by enforcing mandatory back-linking, which ensures every mention of a person or company links back from the entity's page to the mentioning page. This maintains graph integrity during the auto-link reconciliation process.

Does context grounding require a read-before-external-call constraint for meetings and citations?

Context grounding requires a read-before-external-call constraint to ensure the agent searches the brain context store for people, companies, or topics before performing external research. This deterministic behavior contract applies to inbound messages, meetings, and citations.

What is the best way to apply source attribution when writing new facts back to a context store?

The best way to apply source attribution is through deterministic tool usage that writes new information back to the internal context store with proper citations. This ensures every enriched fact is tracked to its original inbound signal or external source.

Can I use entity detection and page loading for continuous background reconciliation of inbound signals?

Yes, entity detection and page loading run as part of an ambient enrichment loop that continuously processes inbound signals. This allows background graph reconciliation and timeline context updates without requiring manual intervention or interrupting user interactions.