BKMRK is an AI-driven bookmarking tool that analyzes content like YouTube transcripts and GitHub repos to score their relevance to your specific coding projects.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install bkmrk
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install bkmrk 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 BKMRK?
BKMRK transforms static bookmarks into an intelligent library of actionable insights. By connecting Openclaw Skills to your BKMRK account, you can leverage Claude AI to perform deep content extraction across diverse sources including X articles, YouTube videos, and research papers. This skill evaluates every piece of saved content against your defined tech stacks and project focus areas, providing implementation suggestions and relevance scores to help you stay focused on what matters most for your development work.
The system uses a structured pipeline—moving from new to staged to done—allowing developers to triage high-volume information feeds into a manageable task list. Because it integrates directly with Openclaw Skills, your AI agent can automatically process new information and suggest how it applies to your current codebase.
BKMRK Use Cases
- Triage high-value learning materials by filtering for relevance scores above a specific threshold.
- Automatically extract and analyze transcripts from long-form technical podcasts or YouTube tutorials to find code snippets.
- Manage a development learning pipeline by moving bookmarks through defined status stages.
- Search across an entire analyzed library for specific technical solutions, authors, or implementation patterns.
How BKMRK Works
- The user connects their BKMRK library using an API key which allows the Openclaw Skills integration to access stored data.
- Coding projects are defined within the system, including specific tech stacks (e.g., React, Node.js) and custom analysis personas.
- Content is submitted via URL or synced from social platforms, triggering deep extraction of text, READMEs, or transcripts.
- Claude AI analyzes the content against the user's active projects to generate a relevance score and implementation suggestions.
- The agent retrieves these insights to help the user decide whether to stage the item for immediate action or archive it.
BKMRK Setup
To get started, obtain your API key from the BKMRK settings page at https://bkmrkapp.com/settings. If you do not have an account, you can use the onboarding endpoint to generate a key immediately:
curl -X POST https://bkmrkapp.com/api/agent/onboard \
-H "Content-Type: application/json" \
-d '{"email": "[email protected]", "consent": true}'
Once you have your key, configure it as the X-API-Key header in your Openclaw Skills environment to begin managing your bookmark intelligence.
BKMRK Data Schema & Taxonomy
The skill organizes data around Projects and Bookmarks to provide a contextual research environment. Data is structured as follows:
| Entity | Key Metadata |
|---|---|
| Project | UUID, Tech Stack, Focus Areas, Analysis Persona, Scoring Bias |
| Bookmark | UUID, Source (Tweet, YouTube, Web), Content Body, Relevance Score |
| Pipeline Status | new, staged, done, trashed |
| Analysis | Per-project implementation prompts and executive summaries |
All library queries can be filtered by project_id, min_score, and priority to maintain a high signal-to-noise ratio.
BKMRK Advanced Features
- Custom Analysis Personas: Inject specific roles (e.g., "Senior iOS Developer") into the AI prompt for domain-aware analysis.
- Deep Re-analysis: Trigger Claude Sonnet to perform intensive project-specific reviews of existing bookmarks.
- Batch Processing: Update the status of multiple bookmarks simultaneously to maintain a clean library.
- Uncapped Extraction: Process 2-hour podcast transcripts and full X Article bodies without content truncation.
SKILL.md
Loading
Related Openclaw Skills
Matrix Channel Repair (matrix-fix)
A specialized maintenance skill for diagnosing and repairing Matrix communication channel issues within Openclaw Skills environments.
Xiaohongshu Data API
A specialized skill for programmatically fetching public Xiaohongshu posts, trending topics, and user data via the TikHub API.
V2Ray Proxy Manager
An automated utility for managing V2Ray proxy services and system-wide proxy settings via the command line.
Super Self-Improving Agent
An advanced AI agent skill that uses multi-modal memory and feedback loops to continuously optimize performance through meta-learning.
TOON Context Compression
A specialized CLI utility that transforms JSON data into the TOON format to achieve approximately 40% savings in token context usage.
charger
A sophisticated CLI tool for monitoring electric vehicle charging station availability via Google Places data.
METADATA
Featured*
