Meridian is a strategic AI intelligence agent that converts fragmented industry noise into structured, timeline-based narratives for product leaders.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install meridian-intel
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 meridian-intel 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 Meridian?
Meridian serves as a high-fidelity intelligence assistant specifically designed for product managers who need to navigate the rapidly evolving AI landscape without suffering from information overload. By moving beyond simple link dumps, this tool synthesizes real-time data into cohesive stories that highlight the cause-and-effect relationship between industry events. It operates as part of the Openclaw Skills ecosystem to ensure that every piece of intelligence is rooted in verifiable facts and temporal context.
The core value proposition of Meridian lies in its ability to answer the so what for product leaders. It doesn't just report that a new model was released; it connects that release to broader technical shifts, business funding, and community sentiment. By enforcing a strict no-source, no-fact policy, it provides a reliable foundation for high-stakes product decision-making.
Meridian Use Cases
- Weekly landscape scanning to identify breakout tools, platforms, and regulatory shifts in the AI sector.
- Deep-dive entity tracking to monitor the specific activities, interviews, and keynotes of industry leaders and competitors.
- Ecosystem product discovery to find trending plugins, high-growth GitHub repositories, and breakout projects within platforms like Openclaw Skills.
- Generating executive-ready intelligence briefings that connect technical milestones to market opportunities.
How Meridian Works
- The agent identifies the intent from the user query, selecting between Broad Scan, Entity Tracking, or Product Discovery modes.
- It establishes a strict temporal window, defaulting to the last 7 days to ensure maximum relevance and information freshness.
- The skill executes multi-channel searches across the web, HackerNews API, GitHub Search API, and arXiv using date-constrained queries.
- Raw signals are filtered based on quality metrics, such as HackerNews points (minimum 30) or GitHub star velocity.
- High-value results are processed via deep browser reading to extract specific quotes, dates, and causality chains.
- Fragmented events are woven into a chronological narrative, complete with actionable PM insights and direct source links.
Meridian Setup
To deploy Meridian within your environment, ensure your agent has access to the required search tools and the local filesystem for report generation. As a part of the Openclaw Skills library, configuration is handled through the standard skill manifest.
# Standard setup involves defining the intelligence directory
mkdir -p intel/briefing intel/entity intel/discovery
# Ensure your search providers (Web, GitHub, HN) are authenticated
openclaw configure meridian
Meridian Data Schema & Taxonomy
Meridian maintains a rigorous data structure to ensure intelligence is easy to navigate and audit. It uses specific templates located in the references directory to generate its outputs.
| Data Type | Target Path | Key Metadata |
|---|---|---|
| Intelligence Briefings | intel/briefing_{date}.md |
Category, Timeline, PM Insight |
| Entity Reports | intel/entity_{name}_{date}.md |
Source URLs, Quotes, Event Clusters |
| Product Discovery | intel/discovery_{platform}_{date}.md |
Star Counts, Downloads, Community Discussion |
| Storyline Narratives | references/storyline-template.md |
Causality Links, Risk Assessment |
Meridian Advanced Features
- Multi-source cross-verification requiring a minimum of two independent sources to validate a storyline node.
- Automated signal-to-noise filtering using community engagement metrics from HackerNews and GitHub.
- Anti-hallucination protocol that strictly prohibits the generation of any URL or fact not found in the search results.
- Actionable insight generation that translates technical updates into strategic product advice.
- Contextual entity tracking that monitors specific X/Twitter and YouTube channels for primary-source announcements.
SKILL.md
Loading
Related Openclaw Skills
Brand Monitoring
A proactive brand monitoring skill to detect trademark infringement, impersonation, and brand abuse across digital channels.
SEO Technical: Sitemap
A technical SEO skill designed to automate the creation, auditing, and optimization of XML sitemaps for maximum search engine visibility.
robots-txt
A technical SEO skill for configuring, auditing, and optimizing robots.txt files to manage search engine visibility and AI crawler access.
FakeX-Till-You-AI
An intelligent posting layer that converts daily AI digests into personalized X post drafts with automated or semi-automated scheduling options.
Quant Stock Picker
An automated AI quantitative stock selection system that analyzes A-share markets using a multi-dimensional scoring model to generate daily investment reports.
Guofeng Lyric Writer
A specialized AI assistant for composing authentic traditional Chinese-style (Guofeng) lyrics with classical imagery and rhyme schemes.
Featured*
