Sombra AI Reader Mode for Openclaw

Sombra provides a persistent research library for AI agents to save, organize, and distill web content into clean markdown context.

dazld
v1.0.1
Mar 21, 2026
0
720
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install sombra-ai-reader-mode

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install sombra-ai-reader-mode using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Sombra AI Reader Mode?

Sombra is a specialized MCP server designed to give your AI agent a persistent research library that lasts between sessions. By integrating Sombra with your Openclaw Skills, you enable your agent to read from a curated library of web pages, notes, and distilled context rather than starting every task from scratch.

This skill allows agents to save permanent copies of web pages as pristine markdown, stripping away advertisements, navigation bars, and layout clutter. It solves the problem of agent drift by providing a source of truth that stays relevant and accessible across various project lifecycles.

Sombra AI Reader Mode Use Cases

  • Building a persistent research library for long-term software projects.
  • Distilling complex technical documentation into dense, code-heavy context summaries.
  • Preventing AI agent drift by providing verified, current documentation as a primary source of truth.
  • Organizing project-specific web discoveries into manageable collections for future sessions.

How Sombra AI Reader Mode Works

  1. The user or agent triggers a URL save, which Sombra fetches and converts into clean markdown.
  2. Content is organized into collections that are scoped to specific projects or technical tasks.
  3. The agent utilizes built-in prompts to distill these collections into high-density context summaries while preserving code blocks.
  4. All saved artifacts and collections are served via the MCP server protocol directly to the agent's context window.
  5. Version history tracks every change, allowing users to roll back any unintended agent-driven modifications.

Sombra AI Reader Mode Setup

To integrate this with your Openclaw Skills, follow these steps:

  1. Sign up at sombra.so and generate a token in Settings > Access Tokens.
  2. Configure your environment by adding the server to your configuration file:
npx -y sombra-mcp --token sombra_pat_YOUR_TOKEN_HERE
  1. Ensure your agent has permissions to access the sombra-mcp server to enable persistent knowledge management.

Sombra AI Reader Mode Data Schema & Taxonomy

Sombra organizes information using a structured taxonomy optimized for AI retrieval:

Component Description Key Metadata
Artifacts Clean markdown versions of web pages or notes Title, Content, URL, Version History
Collections Scoped groups of artifacts for specific tasks Collection ID, Description, Distilled Context
Resources Exposed URI schemes for direct agent access sombra://project/{id}
Archive A recovery layer for deleted content Deleted timestamp, Original collection ID

Sombra AI Reader Mode Advanced Features

  • Context Engineering: Specialized prompts for technical distillation that preserve code syntax verbatim.
  • Version Control: Complete history of every artifact change to prevent data loss during agent rewrites.
  • Bulk Management: Support for moving and updating multiple artifacts simultaneously across collections.
  • Searchable Metadata: Deep search capabilities across titles, content, and original URLs within the research library.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*