I gave my blog an agent: auto topic ideas every week

Описание и применение

Автоматический агент генерирует идеи тем для блога каждую неделю, читая опубликованные посты и предлагая пять новых тем, используя GitHub Actions и LLM DeepSeek.

Задача

Choosing blog topics was time‑consuming and guesswork.

Результат

The agent produced five relevant topics and a scheduling plan on its first run, outperforming the author's intuition.

Рабочий процесс

• Schedule GitHub Action to run weekly
• Read RSS feed to list existing post titles
• Send user prompt to DeepSeek model with tool function
• Model calls tool to fetch titles
• Agent loops until no more tool calls
• Return generated topics
• Write topics to weekly/<date>.md and commit

Доказательства

On its first run it read 17 published posts and replied (excerpt): The existing content is mostly low-level build / framework internals, but "running an agent on an edge device" is almost a blank spot.

Источник

Model is DeepSeek (OpenAI-compatible). The loop: ...

Источник