Cron Scheduled Ollama Autonomous Agent

Description and use

An automated agent using Ollama generates content, then passes it through controversy and credit gates before publishing to social channels, reducing harmful content risk and ensuring compliance.

Problem

Automating content generation and publishing while ensuring safety, user verification, and compliance with constraints like latency and cost.

Outcome

The agent now safely publishes content with automated safety and verification checks, reducing risk of harmful posts and ensuring only verified users contribute.

Workflow

• Generate voice profile from user input and store as JSON
• Run content through controversy gate model; quarantine if flagged
• Probe Postgres database for user verification; quarantine if unverified
• Publish verified content to Twitter, Telegram, and Mastodon

Evidence

The agent operates on a strict pipeline: first, it generates a voice profile from a user’s input, stored as a separate JSON file.

Source

The use of a second model for controversy checking increases inference latency and cost.

Source