Automatically ingest X/Twitter bookmarks, filter marketing noise, and extract actionable technical signals into structured intelligence for humans and AI agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install no-cap
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 no-cap 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 no-cap?
no-cap is a powerful ingestion engine built for the Openclaw Skills ecosystem that solves the problem of information overload on social media. It programmatically pulls your recent X/Twitter bookmarks and subjects them to a rigorous three-layer filtering process to separate genuine technical insights from engagement bait and marketing fluff.
By leveraging this skill within the Openclaw Skills framework, users can maintain a high-signal knowledge base without manual curation. The tool produces dual outputs: a Signal Master Markdown file specifically optimized for AI agent context windows, and a professional HTML email digest for human consumption, ensuring that every bookmarked insight is actually put to work.
no-cap Use Cases
- Extracting technical mechanisms and patterns from industry threads while ignoring promotional CTAs.
- Routing specific project-related bookmarks directly into active development workflows.
- Creating a persistent, searchable 'Signal Master' database that AI coding agents can use for project context.
- Automating a daily or weekly email digest of saved resources using the Resend API.
How no-cap Works
- Authentication: The skill uses session cookies (via auto-login or manual input) to securely fetch your latest bookmarks.
- Three-Layer Noise Filter: Every bookmark is evaluated at the account level (reputation), post-level (engagement bait), and content-level (novelty and specificity).
- Intent Classification: Bookmarks are categorized as signals, action items, project-specific data, learning resources, or tools.
- Signal Extraction: The system strips away narrative fluff to isolate the core mechanism or insight, identifying 'signals in the noise' even in promotional posts.
- Structured Output: The skill generates a timestamped session file and updates a cumulative MASTER.md file to serve as a long-term reference.
- Human Digest: If configured, an HTML email is generated and sent via Resend, featuring curated action items and themed signal cards.
no-cap Setup
To begin using this skill in your Openclaw Skills environment, follow these steps:
- Initialize setup:
/no-cap setup
- Authenticate with X. macOS users can use the automated command to extract Chrome cookies:
cd {repoPath} && npx tsx src/cli.ts auto-login
- For Windows/Linux, manually provide the
auth_tokenandct0from your browser's DevTools. - Define your output directory and specific interests or projects you are tracking.
- (Optional) Provide a Resend API key to enable automated email digests.
no-cap Data Schema & Taxonomy
The skill organizes data to be highly readable for both users and AI agents:
| Component | Format | Description |
|---|---|---|
| Session Files | Markdown | Individual run logs containing signals, noise reasons, and stats. |
| Signal Master | Markdown | A cumulative, compressed file (<200 lines) acting as the primary agent reference. |
| Config | JSON | Stores session credentials (chmod 600), project keywords, and interests. |
| State | JSON | Tracks the high-water mark of processed bookmark IDs to prevent duplicates. |
no-cap Advanced Features
- Aggressive Signal-in-the-Noise extraction that salvages useful ideas from otherwise noisy marketing threads.
- Automated project routing that uses fuzzy matching to link bookmarks to your active development projects.
- Cumulative noise pattern recognition that tracks and auto-filters repetitive low-value accounts over time.
- Agent-first architecture designed to keep the MASTER.md file within efficient LLM context limits.
- Cross-client compatible HTML email templates with automatic link propagation for original sources.
SKILL.md
Loading
Related Openclaw Skills
Context Brief
Context Brief is a specialized memory management tool that enables Openclaw Skills to persist critical session context and decisions across long-running conversations.
Printavo
A powerful integration for Printavo that enables AI agents to automate shop management, order tracking, and customer communications.
LessAccounting AI Agent Skill
An intelligent integration for LessAccounting that enables AI agents to manage invoices, expenses, and financial reports through the Membrane framework.
Open Food Facts
A barcode-driven food database integration for retrieving real-time nutritional facts and product classifications.
Taylor Swift Knowledge Explorer
An advanced research skill for AI agents to analyze the brand history, market impact, and strategic positioning of Taylor Swift as an industry powerhouse.
Multi Search Engine
An integrated multi-engine search skill providing access to 17 global and domestic search platforms without requiring API keys.
Featured*
