Automate a weekly competitor scan that keeps learning
A product manager had Hermes generate a recurring competitor briefing that got quicker and uncovered patterns he had missed.
- What they did
- A product manager set Hermes up for competitive tracking on a Tuesday, then reused the same Monday prompt for six straight weeks. Hermes also checked its own session after every 15 tool calls and wrote a workflow file into its skills folder, which the product manager could open, adjust, or remove. That briefing skill was rewritten four times along the way.
- How it went
- The time needed for the job fell from 20 minutes during the first week to 12 minutes in week four and then 8 minutes by week six. By week six, the briefing was flagging competitor patterns that he had not noticed in three weeks of manual checking.
- Tips
- Keep the saved workflow files under review so you can see what the agent is changing. If you adapt this for a live process, build in approval before any step that would submit, delete, send, or pay for anything.
Try it yourself with Hermes Agent
Set up a recurring [day of week] briefing for [competitor names and URLs]. Each time, review their recent blogs, release notes, pricing pages, hiring posts, and press coverage from the last [time period]. Summarize the biggest changes, explain the likely impact on [my product], and note any repeated signals across weeks. After the first run, write a short skill note about what worked and what should change next time, then show me the draft and wait for my approval before you schedule the recurring job or do anything irreversible.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Turn voice memos into structured Obsidian notes
Have an agent convert spoken Telegram messages into Obsidian notes with a transcript, a tighter version, linked context, and a draft.
Let Cursor audit and summarize an Obsidian vault
Beatriz used Cursor to inspect her Obsidian vault, flag structure issues, answer note questions, and log what other sessions changed.
Test Cursor on a real data science task
A data scientist graded Cursor on a Python football project and found it strongest for quick prototypes, but weak on pandas, cleaning, and scraping.
Convert YouTube videos into linked research notes
Claude Code can gather YouTube videos into NotebookLM, then export cited notes, a graph, audio, and flashcards into Obsidian.
Build a dashboard from a semantic data layer
Use a coding agent with Graphene to define metrics, query your warehouse, and publish a report with charts and screenshots.
Create a multi-agent analysis pipeline in Claude Code
A data analyst had Claude Code assemble a five-part analysis system from a detailed specification.