Moxie Docs: Knowledgebases

The Friday Cleanup idea is nicely restrained. Batching doc drift into one reviewable PR, with nothing auto-merging, feels like a much easier habit for a team to trust.

the code-vs-docs source-of-truth question got covered for the bug scenario, but there's a different case worth asking about: forward-looking design docs or RFCs describing a feature that's planned but not built yet. since code is ground truth for drift detection, does Moxie treat a doc describing not-yet-implemented behavior as "drift" and try to flag or rewrite it toward what the code currently does, or is there a way to mark a doc as intentionally ahead of the code so it doesn't get flattened into just describing the present state?

Congrats on the launch. Living docs plus an MCP server for the same repo is a smart pairing, most tools solve doc-drift or agent-context but not both from one source of truth. Before I'd point Public Sites at a client's docs: is there a filter step before publishing that catches anything meant to stay internal, a staging URL or an internal-only heading written into the markdown, or does that rely on the team keeping the source clean themselves?

The PR check is the real forcing function here. Quick question—when an agent hits undocumented code through the MCP, does it try to fill the gap or just work around it? That behavior probably determines whether this compounds your docs problem or solves it.

The PR check is the part I would lead with. Docs rot because nothing in the workflow ever tells you they rotted, so a check that fails loudly on the diff is worth more than any amount of generation quality.One thing worth thinking about on the MCP side: how the docs get chunked matters as much as how they get written. An agent pulling repo context does much better with small single-topic sections it can retrieve precisely than with one long well written page, because a long page embeds as an average of everything in it and stops matching specific questions.Congrats on the second launch.

the PR check that keeps docs honest is the part that would actually get me to adopt this - every docs tool I've tried eventually goes stale because nothing forces the update at merge time. does it block the PR outright when docs drift too far from the code change, or just flag it as a warning someone can ignore?

Is this a kind of protection against accidental changes made by the AI?

This feels useful because docs are starting to serve two audiences at once: humans reading the public knowledgebase, and AI agents using repo context while work is happening.The trust boundary I’d care about is the step before something becomes public. Temporary implementation notes are often helpful for an agent, but not always safe or polished enough for users. A clear diff/review queue for “agent-found doc drift” before publishing would make this much easier to trust.

This is interesting. What's the indexing latency on a large monorepo with frequent commits?

Congrats on the launch! On the PR checks that keep docs honest, does it block the merge if docs go stale, or just leave a comment flagging what needs updating?

Really keen to test this out and see how it differs from using pre-commit documentation update hooks, which is my current setup. One quick question for ya - how do you handle the case where the docs are right and the code is wrong? On a bad merge, "sync to code" would happily overwrite the doc that describes the intended behavior. Does drift detection ever flag the code side, or is source always treated as ground truth?

Fantastic idea! Wondering if there is a document update every time the repo content changes? Does it run periodically? or is it all one and done (I doubt it!)

@cadenjs - love the idea, but don't like the subscription model too much. What if you only need this for 1 repo and the docs are only typically updated 2 to 3 times a year? Paying a monthly fee seems excessive, are you not considering a pay as you go model for smaller projects? Perhaps something tiered based on usage volume rather than a flat monthly/annual subscription.

What could you build with GPT-5.6 that wasn’t practical before?

Hey all!We've greatly expanded and improved our developer documentation & MCP tooling - giving internal devs useful docs that are always up to date, and tools for their AI agents that reduce token usage and improve code quality output. We've build a public open-source MCP server, a CLI setup tool, and a prompt to get setup easily in 1-paste.The biggest ask we've seen was: How do we manage public-facing docs? Help centers? Knowledgebases?Now you can! - https://moxiedocs.app/ Starting in Beta any Moxie Docs user can opt-in themselves to start using our Public Sites feature. This unlocks:Automatically generated documentation for both developers and consumers targeted for public consumption. Rich with layouts, charts, codeblocks, internal linking, and more.White-labelling - add your logo, brand color, and light instruction changesComing soon custom domains and remove Moxie Docs linking for Pro or higher plansFully hosted, you control what documentation is published, all exists in your code base no vendor lock-in, you own your data. Always.Full document search, reading time, reading indicator, table of contents, documentation groupinIntegrates directly with our Changelog feature - get a hosted changelog directly linkable and tied to your help centerCheck it out with our own dogfooding example 👉https://docs.moxiedocs.app/What else has shipped?Webhooks integration - set up any custom webhook endpoint to get alerted when moxie docs finishes indexing, PRs are opened, or Friday Recap is ready for reviewSlack integration - set up the same alerts and summaries to any Slack workspaceImproved UI / UX for pagination, bulk actions, and moreFaster loading times app-wide, and an upgrade to our DB and infrastructure to improve performanceSee what else is new on our hosted Moxie Docs changelog: https://docs.moxiedocs.app/changelog