digitallyinduced

digitallyinduced/ihp

Haskell MIT Web Dev

🔥 The fastest way to build type safe web apps. IHP is a new batteries-included web framework optimized for longterm productivity and programmer happiness

5.3k stars
223 forks
active
GitHub +6 / week

5.3k

Stars

223

Forks

261

Open issues

30

Contributors

v1.6.0 21 Jun 2026

AI Analysis

IHP is a batteries-included web framework built on Haskell and Nix that emphasizes type safety, developer productivity, and AI-assisted development workflows. It is purpose-built for teams developing serious web applications who prioritize compile-time error detection and long-term maintainability over rapid prototyping. This framework is best suited for developers with Haskell experience or teams willing to adopt a typed functional approach; it is not a general-purpose framework for JavaScri...

Web Dev Web Framework Discovery value: 4/10
Documentation 8/10
Activity 9/10
Community 7/10
Code quality 8/10

Inferred from signals mentioned in the README (tests, CI, type safety) — not a review of the actual code.

Overall score 8/10

AI's overall editorial judgment — not an average of the bars above, can weigh other factors too.

type-safe-web-framework haskell-based ai-assisted-development nix-managed-environment functional-programming
Actively maintained Well documented MIT licensed Niche/specialized use case Production ready
Deep Analysis · Based on README and public signals
2w ago

IHP brings Rails-style productivity to Haskell web development with type-safety and Nix-managed environments

IHP (Integrated Haskell Platform) is a batteries-included web framework targeting teams who want the productivity of a convention-over-configuration framework with the compile-time safety of Haskell. It ships with code generators, a visual schema designer, built-in authentication, real-time data sync, and a Nix-managed dev environment. It targets Haskell-fluent developers and small product teams willing to accept Haskell's learning curve in exchange for stronger correctness guarantees. In production since 2017 according to its own README, it appears to serve a genuine but narrow audience.

Origin

Created in December 2017 by digitallyinduced, a German software consultancy. Evolved from an internal productivity tool into a public framework. Has grown steadily rather than virally, accumulating ~5,300 stars over roughly 8 years.

Growth

Growth has been slow and steady, reflecting the natural ceiling of the Haskell developer population. Interest appears driven by Haskell enthusiasts seeking a practical full-stack option rather than broader web development trends. Recent addition of AI-assisted development features (CLAUDE.md, OpenAI integration) suggests an attempt to broaden appeal by positioning the type system as an AI safety net.

In production

README states 'In production since 2017. Used by teams building serious web applications.' digitallyinduced appears to use it for client projects internally, providing some commercial backing. Independent third-party production usage is not independently verified from available metadata, though the 8-year track record and active forum/Slack community suggest non-trivial real usage beyond the maintainer itself.

Code analysis
Architecture

Appears to follow an MVC pattern with type-safe routing, controllers, and HSX (JSX-like) templating. Likely built on GHC extensions for type-level programming. Monorepo structure with clearly separated packages (ihp-datasync, ihp-graphql, ihp-ssc, etc.) suggests modular design. Nix is used for reproducible environments, and PostgreSQL is the assumed database. Deployment model is NixOS-native with systemd socket activation for zero-downtime deploys.

Tests

An ihp-hspec package is listed in the ecosystem table, indicating first-party testing utilities exist. Actual coverage levels are not documented in README.

Maintenance

Last push was 2026-06-23, four days before the evaluation date — clearly actively maintained. The project has been sustained for over 8 years, suggesting long-term commitment from digitallyinduced as a commercial backer. Recent README updates referencing Claude Code and AI workflows indicate the project is tracking current developer tooling trends.

Honest verdict

ADOPT IF: your team already knows Haskell and values compile-time safety for a long-lived web application, or you are willing to invest in Haskell's learning curve for a product that needs strong correctness guarantees. AVOID IF: your team lacks Haskell experience and cannot afford the ramp-up, if you need a large third-party plugin ecosystem, or if fast hiring for the stack is a business requirement. MONITOR IF: you are interested in how AI-assisted development interacts with strong type systems, as IHP's positioning here may prove prescient or may fail to attract meaningful new adoption.

Independent dimensions

Mainstream potential

2/10

Technical importance

7/10

Adoption evidence

3/10

Risks
  • Haskell's relatively small developer population creates a permanent ceiling on adoption and makes hiring for IHP-based projects difficult.
  • Strong dependency on Nix and NixOS for the intended deployment path may be a barrier for teams with existing infrastructure constraints or Kubernetes-centric workflows.
  • digitallyinduced is a single commercial entity; if the company changes direction, long-term maintenance continuity is uncertain despite the MIT license.
  • The Haskell ecosystem can have slow or breaking GHC upgrade cycles, and IHP's reliance on specific GHC extensions makes it potentially vulnerable to upstream churn.
  • The 'AI-assisted development' positioning appears opportunistic relative to the current date and may not meaningfully differentiate IHP from other strongly-typed frameworks in practice.
Prediction

IHP will likely remain a well-maintained, niche framework serving a stable but small audience of Haskell-committed teams. Mainstream breakout appears unlikely given Haskell's structural hiring constraints, but the project shows no signs of abandonment.

0 found this helpful

Newsletter

Get analyses like this every Monday

Free weekly digest of the most interesting open-source discoveries.

Languages

Haskell
89.9%
Nix
4%
TypeScript
1.9%
JavaScript
1.8%
HTML
0.9%
CSS
0.9%
Shell
0.4%
Python
0.2%

Information

Language
Haskell
License
MIT
Last updated
1d ago
Created
105mo ago
Analyzed with
anthropic/claude-haiku-4-5

Stars over time

Loading…

Contributors over time

Top 100 contributors only — repos with more will plateau at 100.

Loading…

Similar repos

hapijs

hapijs/hapi

Hapi is a mature Node.js web framework designed for building scalable, secure...

14.8k JavaScript Web Dev
h3js

h3js/h3

H3 is a minimal HTTP framework for TypeScript/JavaScript built for high...

5.4k TypeScript Web Dev
dmjio

dmjio/miso

Miso is a Haskell web framework for building web and mobile applications using...

2.4k Haskell Web Dev
iii-hq

iii-hq/iii

iii is a runtime platform that enables services to dynamically discover,...

18.4k Rust Dev Tools
facebook

facebook/hhvm

HHVM is an open-source JIT-compiled virtual machine for executing programs...

18.6k C++ Web Dev
vs. alternatives
Yesod (Haskell)

The established alternative Haskell web framework. IHP differentiates by offering higher-level conventions, code generation, and a visual IDE, targeting developers who want less boilerplate than Yesod's more explicit approach.

Ruby on Rails

IHP is explicitly modeled on the Rails productivity philosophy but trades Ruby's dynamism for Haskell's type safety. Rails has orders of magnitude more adoption and ecosystem maturity.

Phoenix (Elixir)

A closer philosophical peer — both target functional-language web development with strong conventions. Phoenix has significantly broader adoption and a larger ecosystem, partly because Elixir has a lower barrier to entry than Haskell.

Servant (Haskell)

Servant focuses narrowly on type-safe APIs rather than full-stack web apps. IHP covers more ground but is correspondingly more opinionated. They target somewhat different use cases.

Next.js / full-stack JS

Represents the mainstream alternative for teams prioritizing time-to-hire and ecosystem size over type-level guarantees. IHP competes on correctness and long-term maintainability, not on talent availability.