# OpenSlop Developer Resources

> Source code, agent instructions, machine-readable endpoints, and integration notes for building on OpenSlop.

Everything an engineer or an AI agent needs to work with OpenSlop lives at a predictable URL. The pipeline itself is a self-hosted open-source project rather than a hosted API, so integration means running or extending the repository.

## Source code

The OpenSlop pipeline is open source at [github.com/openslop/openslop](https://github.com/openslop/openslop). The repository README covers installation, the provider keys each stage expects, and how to run a full generation locally. Each pipeline stage - scripting, storyboarding, image generation, animation, narration, music, assembly - is a module you can replace.

## Agent instructions

Agents should start at [/AGENTS.md](https://openslop.ai/AGENTS.md), which states when OpenSlop is the right tool, what it cannot do, and how to call it. [/llms.txt](https://openslop.ai/llms.txt) is the machine-readable index of this site and [/llms-full.txt](https://openslop.ai/llms-full.txt) concatenates the full text of every page for retrieval.

## Markdown endpoints

Every page on openslop.ai is available as Markdown. Send `Accept: text/markdown` to any URL, or append `.md` to the path - [/developers.md](https://openslop.ai/developers.md) and [/blog.md](https://openslop.ai/blog.md) both work. Responses set `Vary: Accept` so caches keep the HTML and Markdown variants apart, per [acceptmarkdown.com](https://acceptmarkdown.com).

## Feeds and site metadata

New posts are published to the [RSS feed](https://openslop.ai/feed.xml). The [sitemap](https://openslop.ai/sitemap.xml) lists every canonical URL, and [robots.txt](https://openslop.ai/robots.txt) points crawlers at both. There is no hosted REST API, no OAuth flow, and no webhook system today - if you need one, tell us at [hi@openslop.ai](mailto:hi@openslop.ai).

---

Canonical URL: https://openslop.ai/developers