KaggaBot: Automating Daily Verse Posts With GitHub Actions
Table of Contents
What’s Mankuthimmana Kagga #
Mankuthimmana Kagga — often called the Kannada Gita — is a collection of 945 philosophical verses by D.V. Gundappa (DVG). Written in the 1940s, it wrestles with life’s biggest questions with a rare combination of humility and depth. It deserves a wider audience.
KaggaBot #
It’s a small automation that does one simple thing: post two verses from the Kagga every day to Mastodon (an ad-free social media platform), automatically, without anyone pressing a button.
What’s in the code #
The core of the bot is a Python file containing all the verses of the Kagga, individually scraped and stored as structured data.
A script picks from this collection and formats each verse for posting. GitHub Actions handles the scheduling — no server required, no ongoing cost.
Verses in .py file → bot script → GitHub Actions cron → Mastodon post
The scheduling magic #
A GitHub Actions workflow file defines two daily triggers — morning and evening. When the schedule fires, GitHub spins up a clean environment, checks out the code, and runs the bot. That’s the entire pipeline.
Two posts a day, every day #
on: schedule: - cron: ‘0 6 * * *’ # morning - cron: ‘0 18 * * *’ # evening No cloud bill. No server to maintain. No babysitting.
Why Mastodon? #
Mastodon is open, ad-free, and has a clean API — ideal for a bot that exists purely to share literature, not to game an algorithm. The posts are public and readable by anyone, no account needed.
What Broke #
GitHub Actions cron doesn’t mean “exactly at this time.” The workflow is scheduled for 6am and 6pm, but GitHub runs scheduled workflows when capacity is available — which during busy periods can mean a delay of 15–30 minutes. For a poetry bot, that’s fine. For anything time-sensitive, it would be a problem worth knowing about upfront.
The ephemeral environment forgets everything. GitHub Actions spins up a clean runner for each job. There’s no persistent file system between runs. This matters for state tracking — knowing which verse was posted last. The solution is committing the current verse index back to the repository after each run. If you forget this, the bot picks a verse randomly (or always posts verse #1) on every run. First version did exactly that.
GitHub disables scheduled workflows on inactive repos. If a repository has no activity for 60 days, GitHub automatically disables scheduled workflows. The bot will simply stop posting with no notification. The fix is either a periodic commit to keep the repo active, or enabling the workflow manually when it gets disabled. Worth knowing before you rely on it as a daily habit.
The Mastodon API token in GitHub Secrets. Storing the token as a repository secret works, but if the secret is misconfigured (wrong variable name in the workflow YAML) the bot fails silently — no post, no obvious error in the logs unless you know to look at the environment variable step. Naming the secret MASTODON_ACCESS_TOKEN and double-checking the ${{ secrets.MASTODON_ACCESS_TOKEN }} reference in the YAML saved a lot of confusion.
Why this matters #
At two posts a day, it takes over a year to work through all 945 verses of Mankuthimmana Kagga.
That’s the right pace for philosophy — slow enough to sit with, regular enough to build a habit. KaggaBot transforms reading a classic Kannada text into a daily practice, entirely on autopilot.
The whole project is open source. If you want to adapt it — for a different text, a different language, a different platform — the bones are all there.