Hermes Agent

Blogizi for Hermes Agent

Nous Research's open-source agent framework can run blogizi-agent directly — including scheduled and recurring drafting tasks defined in config.

View Docs

Blogizi drafts by default — publishing live always requires explicit confirmation.

Context

What is Hermes Agent?

Hermes Agent is an open-source AI agent framework by Nous Research. It is tool-execution based: the agent discovers skills and runs commands against tools you enable. It works with multiple LLM backends — Blogizi plugs in as a skill via the bundled SKILL.md.

Install

Add the Blogizi skill

Install once, set your account API key, then confirm Hermes can see the tool.

Terminal

npx skills add bytemindlab/blogizi-agent
export BLOGIZI_API_KEY=your_api_key
hermes tools list

Scheduled drafting

Recurring drafts via hermes-config.yaml

Hermes’ config-driven tasks are the differentiator — set a schedule so the agent drafts on a cadence without you kicking it off each week.

hermes-config.yaml

tasks:
  - name: weekly-blog-draft
    schedule: "0 8 * * 1"
    description: >
      Check ~/content/pending for new post ideas.
      Write a draft with SEO frontmatter and push it
      to Blogizi as a draft. Do not publish — leave it
      for review in the dashboard.
    tools:
      - blogizi

Scheduled tasks should call blogizi draft, not blogizi publish. The recurring job pushes content as a draft for review. Going live still requires a human to review the post and run blogizi publish (or confirm publish in the dashboard) — that is intentional safety, not a missing feature.

Fit

Why Hermes + Blogizi

Open-source and inspectable
Tool-execution framework runs blogizi-agent commands directly
Scheduled tasks — recurring drafts without a manual trigger each time
Works with any LLM backend Hermes supports
Draft-first — scheduled runs never publish without explicit confirmation

Schedule the draft. You decide what goes live.

Hermes can run Blogizi on a cron. Publishing stays an explicit step after you review the draft.

Full agent docs

CLI, OpenClaw, Claude Code, and MCP overview: /agent