cercano-init

Scan a repository and write a .cercano/context.md file.

Updated Mar 20, 2026
One-click install
npx skills add https://github.com/GoogleDevRelExplorations/cercano --skill cercano-init-googledevrelexplorations
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cercano-init
Source: https://github.com/GoogleDevRelExplorations/cercano/tree/main/.agents/skills/cercano-init
Command: npx skills add https://github.com/GoogleDevRelExplorations/cercano --skill cercano-init-googledevrelexplorations

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Scans a repository to generate a centralized, up-to-date .cercano/context.md, enabling Cercano tools to operate with project-specific knowledge without manual setup.

Core Features & Use Cases

  • Scans key files and project structure to extract domain knowledge and compile it into a reusable context document.
  • Merges discovered information with any host-provided context and prepends it to future Cercano tool calls in the project.
  • Onboard new projects quickly by creating a ready-to-use knowledge base that improves local model responses.

Quick Start

Run this init once per project to generate the .cercano/context.md and enable context-aware Cercano tooling.

Frequently Asked Questions about cercano-init

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

FAQPage Schema
How do I generate project-scoped context for local AI tools from a repository?

Scanning a repository to generate project-scoped context involves extracting domain knowledge from key files and project structure, then writing it to a .cercano/context.md document. This file enables local AI tools to operate with project-specific awareness.

What is repository onboarding for local AI models and when do I need it?

Repository onboarding for local AI is the process of creating a centralized knowledge base from your codebase. You need it when initializing new projects or after major repository changes to improve local model response accuracy.

Do I need a running Ollama server to initialize project context?

Yes, initializing project context requires a running Cercano server with Ollama available locally. The process reads repository contents and merges them with host context to generate the final context document.

How do I onboard a new project to use local AI with Cercano?

To onboard a new project, run the init process once per project. It scans key files, extracts domain knowledge, and writes a ready-to-use .cercano/context.md that prepends context to future Cercano tool calls.

What's the best way to maintain consistent domain knowledge for local AI across a project?

The best way to maintain consistent domain knowledge is generating a centralized .cercano/context.md file. By scanning the repository and merging host context, it ensures all Cercano tools share up-to-date project information without manual setup.