Hermes Atlas
Automations & use cases · works with Hermes Agent

CrossPulse

hooper-lee/crosspulse

Daily cross-border e-commerce market monitoring that scores signals and pushes a report to Feishu

In short

CrossPulse is an automated market-monitoring workflow for cross-border e-commerce sellers. It collects news signals for platforms such as Amazon and TikTok Shop, analyzes and scores them, and pushes a daily report to Feishu. The repository is tagged for Hermes Agent.

What CrossPulse does

CrossPulse is an automated market-monitoring workflow for cross-border e-commerce sellers. Each day it collects signals from Google News RSS and industry media such as Retail Dive and eCommerceBytes, covering Amazon, TikTok Shop, Shopify, Temu, SHEIN, eBay and Walmart. Items are deduplicated by URL and title, stored in a Feishu Bitable, analyzed, and turned into a daily report that is pushed to a Feishu group as an interactive card.

Analysis has three modes. The llm mode calls an external model per item and needs an API key, heuristic uses keyword matching and classification rules with no dependencies, and agent lets an AI agent do the analysis itself, which the README rates as the highest quality. Each item gets a Chinese summary, a four-dimension score (importance, trend strength, business value, risk) and a recommendation. The report opens with a red, yellow or green market verdict and lists top opportunity and risk signals, competitor news, early hot-product signs and follow-up items. A DRY_RUN setting tests locally without writing to Feishu.

Key features

  • Collection from Google News RSS and industry media across seven or more platforms
  • Deduplication by URL and title, with storage in a Feishu Bitable
  • Three analysis modes: llm, heuristic and agent
  • Four-dimension scoring of importance, trend strength, business value and risk
  • Daily report pushed to Feishu as an interactive card
  • Streak marking for repeated hot topics and a weekly trend comparison

When to use it

  • Getting a daily briefing on Amazon and TikTok Shop policy changes
  • Spotting opportunity and risk signals for product selection
  • Having an AI agent perform the market analysis in agent mode

Who it is for: Cross-border e-commerce sellers and operations teams, especially those who already use Feishu.

How it fits with Hermes Agent

The repository is tagged for Hermes Agent and has an agent analysis mode, but the README sections available here describe no Hermes-specific setup.

How to install CrossPulse

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

git clone https://github.com/hooper-lee/crosspulse.git
cd crosspulse
pip install -r requirements.txt
cp .env.example .env

Requirements: Python with the packages in requirements.txt, a Feishu (Lark) setup for Bitable storage and card push, and an LLM API key only for llm analysis mode

Note: Reports are delivered to Feishu (Lark), and the full documentation is in Chinese with an English summary.

FAQ

What is CrossPulse?

CrossPulse is a market-monitoring system for cross-border e-commerce that gathers signals from news sources, analyzes and scores them, and sends a daily report to a Feishu group.

Does CrossPulse work with Hermes Agent?

The repository is tagged hermes-agent, and its agent analysis mode lets an AI agent perform the analysis. The README sections available here do not describe a Hermes-specific setup.

How do I install CrossPulse?

Clone the repository, install requirements.txt and copy .env.example to .env with your settings. Then run python3 -m src.run --config config.yaml, or set DRY_RUN=true to test without writing to Feishu.

Similar automations for Hermes Agent

All automations

Related guides: Build agent teams with the Hermes kanban board · How to run Hermes Agent securely