orient

Parse queries and retrieve relevant context from knowledge sources into organized briefings.

9|2|Updated Mar 7, 2026
One-click install
npx skills add https://github.com/harnessprotocol/harness-kit --skill orient-harnessprotocol
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: orient
Source: https://github.com/harnessprotocol/harness-kit/tree/main/plugins/orient/skills/orient
Command: npx skills add https://github.com/harnessprotocol/harness-kit --skill orient-harnessprotocol

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of wasting time by loading or searching too much context when you only need the most relevant information for a specific topic or goal.

Core Features & Use Cases

  • Topic-focused orientation briefings: Produces a structured briefing that surfaces only the most relevant Graph, Knowledge, Journal, and Research items for your query.
  • Multi-source retrieval: Searches MCP Memory Server (when connected), scans knowledge files and journal entries, and lists matching research references.
  • Fast, token-conscious output: Uses strict limits and omits empty sections so you get an orienting overview without a document dump.

Quick Start

Ask the AI to run /orient membrain to produce a focused orientation briefing across your graph, knowledge files, journal entries, and research index.

Frequently Asked Questions about orient

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

FAQPage Schema
How do I get a focused briefing from my knowledge graph and journal entries for a specific topic?

A topic orientation retrieves only the most relevant context from your knowledge graph, journal entries, and research index by parsing your query, performing bounded graph search, and extracting targeted excerpts to produce a structured briefing.

Can I generate a topic briefing if the MCP Memory Server is disconnected?

Yes, you can generate a topic briefing if the MCP Memory Server is disconnected. The retrieval process enforces graceful degradation, scanning available local knowledge files, journal entries, and research references while omitting graph search results from the output.

Does the context retrieval process search across multiple sources like research indexes and local files?

Yes, context retrieval searches across multiple sources including the MCP Memory Server, local knowledge files, journal entries, and the research index. It scans these sources simultaneously to surface matching references and excerpts for your specified topic.

What is the best way to avoid loading too much context when searching for specific entity types and time qualifiers?

The best way to avoid loading too much context is to use a targeted topic orientation. It applies strict limits and omits empty sections, ensuring you receive an orienting overview with only the most relevant information rather than a full document dump.

How do I use conversational commands to extract relevant excerpts by sections from my research index?

To extract relevant excerpts by sections, issue a conversational command specifying your target topics, entity types, or time qualifiers. The system parses the query, searches the research index, and structures the output into an organized briefing with matching excerpts.