Top 200 ReposEXPLORER
Browse agents, categories and pages
Back to results
Reported by sourceThe source says this happened. Top 200 Repos has not reproduced it.

Keep a resident desktop agent for research notes and blog work

Raj Chettri runs Nous Research's Hermes Agent on his desktop for scheduled research summaries, note filing, quick scripts and his blog's writing pipeline.

The problem

How it works

He installed Hermes Agent on his own desktop and relies on its persistent memory, skills and cron scheduling for research summaries, note capture and filing, small scripts and managing his blog's writing pipeline. He describes switching between a local model for sensitive work and a frontier model for heavier jobs, and reaching the agent from his phone.

Results and limitations

He reports that it survives reboots, runs scheduled jobs unattended, and that its memory improved over about a month so he stopped re-explaining project context; he names phone access and output landing in his Obsidian vault as the biggest wins.

Before you try it

No metrics, model names or failure examples are given, and this is the first post in a three-part series. He warns that an agent with shell access on a main machine is a real attack surface and frames his setup as a personal tool, not one for client data.

Tools and repositories

Tools
Hermes Agent · Obsidian · cron
Prerequisites
Needs an always-on desktop and a model provider key or local model.

Try it yourself: Schedule a weekly research digest into my notes

Fill the fields to adapt the prompt. Empty fields stay as visible [placeholders].

Preview
Every [Schedule], collect new items about [Research topic] from these sources only: [Allowed sources]. For each item, write a three-sentence summary in your own words, the link, and why it matters for [Project it should serve]. Save the digest as a new dated note in [Notes folder]; never overwrite or delete existing notes. Before enabling the schedule, run it once, show me the draft note and the cron entry, and wait for my approval. Use [Model for private content] whenever personal notes are involved. Do not post, email or share the digest anywhere, and record each run in [Log file].

Give the agent scoped, rotatable keys and review its logs, since it runs unattended with shell access.

Related videos

About NousResearch/hermes-agent

About NousResearch/hermes-agent

Independent creators. Listing a video implies no endorsement, sponsorship or partnership.

Sources

  1. Meet Hermes: The Self-Improving AI Agent Running on My Desktop (opens in a new tab)

    First-person account · rajchettri-com.pages.dev · Published 25 May 2026 · Source checked 9 Nov 2026, 22:23 UTC

    What this source supports
    • Runs Hermes Agent on his desktop
    • Used for research summaries, note filing, scripting and blog pipeline
    • Runs cron jobs unattended with memory that improved over about a month
    • Warns about shell access as an attack surface
Report a problem

Planned, not done

Hand off research and document tasks to the Dispatch agent in Claude Cowork

Anthropic’s Cowork docs describe Dispatch, a background agent in Claude Desktop that splits one request into child tasks, runs each as its own session and forwards permission prompts to you, even from your phone.

claude.com
Reported by source

Run a self-hosted personal assistant on a spare Mac

Federico Viticci ran OpenClaw (then called Clawdbot) on an M4 Mac mini with Claude Opus 4.5 over Telegram for briefings, smart-home control and small automations.

macstories.netopenclaw/openclaw
Reported by source

Answer technical questions by letting async agents run experiments

Simon Willison sends research questions to cloud coding agents such as Claude Code for web and Codex Cloud, which write and run experiments in a dedicated repo and open a PR.

simonwillison.net · 6 Nov 2025anthropics/claude-code, openai/codex
Reported by source

Build a cited Q&A knowledge base over your own documents

Self-host RAGFlow, upload PDFs, Office files and scans, and ask questions that come back with traceable source chunks.

github.cominfiniflow/ragflow
Reported by source

Build a self-hosted AI agent workflow with human approval

Use n8n's visual canvas to wire an LLM agent to your apps, add tools and memory, and pause for human approval before it acts.

github.comn8n-io/n8n
Reported by source

Build an on-call investigation agent from your own alert history

Goutham Veeramachaneni connected Claude Code to Grafana's gcx CLI, his code and runbooks, then trained its skill on two weeks of past pages for his services.

gouthamve.dev · 21 Apr 2026anthropics/claude-code