Hey Product Hunt! 👋
I built Doctective after wasting one too many hours debugging issues caused by outdated docs. Imagine this: Want to use a new library so you follow the documentation, it doesn't work, and then you discover the code changed 3 months ago but nobody updated the docs.
The worst part? AI coding assistants are now reading these stale docs too, generating broken code based on outdated information.
So I built something simple: connect your GitHub repo, and Doctective watches every PR. When code changes would break your docs, it tells you and can even open a companion PR with the fixes automatically.
No new tools to learn. No manual doc audits. Just accurate docs that update themselves.
Would love your feedback — what's the #1 documentation pain point on your team?
How well does it integrate into existing development workflows (e.g., Git, CI/CD)? For example, can document updates be previewed and reviewed as part of a pull request, rather than modifying the main branch documents directly—only then can true "docs as code" collaboration be achieved.
The claim of "documents that never go outdated" is bold. In actual use, when major code refactoring or deletion occurs, can it intelligently link and update or archive the corresponding document sections, avoiding "document ghosts" that reference invalid content?
For developers, first impressions matter. Is the installation and configuration sufficiently user-friendly? Can it quickly identify the project structure and provide clear, unobtrusive configuration recommendations, instead of forcing me to spend a lot of time adapting to its complex rules?
How do the automatically generated documents fare in terms of readability and practicality? Can it generate truly useful content for developers (even new team members) based on code comments and structure, rather than just a list of functions? Additionally, to what extent can I customize the document style and content focus to align with team specifications?
Great project. Is there an integration with Confluence? So that documentation can be created there automatically?
This solves a very real, very painful problem. Outdated docs now don't just slow teams down, they mislead AI tools too. Tying doc updates directly to PRs and making correctness the default is a smart approach. We've seen a similar pattern while build Dashform, when tools fit naturally into existing workflows, adoption just works. Great launch, excited to follow Doctective!
There’s nothing worse than following outdated documentation and getting errors, so the 'never go stale' part is huge. I wonder if it supports all the major languages right out of the box, or if you have to spend a lot of time configuring it for each project.
Congrats on the launch. This hits a very real pain point, especially now that both humans and AI depend so heavily on docs being accurate. Tying documentation health directly to PRs feels like a practical, low friction way to keep things honest.
This is great! Have you also considered the idea of creating a vscode plugin that will update docs alongside the PRs on commit? Like how huskey has precommit checks, this could essentially verify docs are up to date when pushing the code?
When Doctective proposes doc changes, how do you ensure they’re correct and minimally invasive—what guardrails (diff size limits, source-of-truth selection, tests/build checks, quoting code links) prevent confident but wrong edits from getting merged?
Sounds super useful. What kind of documentation platforms do you support? Do you need to have the documentation in Markdown files in the same repository?
Congrats on the launch — love how Doctective keeps docs in lockstep with every code change.
Auto-fixing docs on every PR is a huge unlock, but at scale the pain is false positives: refactors, renames, and “behavior changes” that AST diffs catch, but the human intent in the docs is nuanced.
Best practice is to ground updates in code-symbol provenance (tree-sitter/LSIF), generate minimal diffs, and run a CI gate that links each doc change to the exact commit and API surface change with an easy approve workflow.
How are you detecting drift today (AST/symbol graph vs heuristic grep), and can Doctective produce deterministic, reviewable PRs that pin the doc update to a specific commit hash and changelog entry?
