01
No pay-to-list
Placement cannot be bought. An entry changes only when someone opens a pull request with evidence — and the diff is public forever.
00 /INDEX
The AI landscape, observed.
An open source observatory of AI tools, stacks and models. No paid listings. Every verdict is a pull request you can audit.
Explore the catalogStar on GitHub
FIELD: AI · OBJECTS: 70 · SHIFT: LIVE
01 /COUNT
- 70
- tools observed
- 7
- stacks mapped
- 20
- models tracked
02 /FEATURED
Bright objects, verified.
The tools we would defend in a code review — marked gold, never paid for.
ChatGPTfeatured
The general-purpose AI assistant that set the category baseline.
writing
freemium · free tier; Plus $20/mo
FREE TIER
shipverified 2026-08-15
Claude Codefeatured
Anthropic's terminal agent that edits, runs and tests code in your repo.
coding
paid · included in Claude Pro from $20/mo; API pay-per-token
API
shipverified 2026-08-15
Cursorfeatured
AI-first code editor (VS Code fork) with agent mode and inline edits.
coding
freemium · from $20/mo (Pro)
FREE TIER
shipverified 2026-08-15
ElevenLabsfeatured
The reference standard for AI voice — TTS, cloning, dubbing and voice agents.
audio
freemium · free tier (10k credits/mo); Starter $6/mo
APIFREE TIER
shipverified 2026-08-15
Midjourneyfeatured
The aesthetic benchmark for AI image generation — Discord-born, now web-first.
image
paid · from $10/mo (Basic)
shipverified 2026-08-15
OpenRouterfeatured
One API for every major model — unified billing, fallbacks and routing.
dev-infra
freemium · pay-per-token pass-through plus a small fee
APIFREE TIER
shipverified 2026-08-15
Runwayfeatured
The incumbent AI video suite — Gen-4 generation plus a full editing toolchain.
video
freemium · free one-time credits; from $15/mo (Standard)
APIFREE TIER
shipverified 2026-08-15
Veo 3featured
Google's video model with native audio generation — the current quality leader.
video
paid · via Google AI Pro $20/mo (limited); API from $0.75/sec
API
shipverified 2026-08-15
03 /LOG
Latest observations.
- 2026-08-15toolAidershipOpen-source pair programming in the terminal — git-native, works with any model.
- 2026-08-15toolAnthropic APIshipClaude models direct — the practitioner's pick for coding and long-form text.
- 2026-08-15toolChatGPTshipThe general-purpose AI assistant that set the category baseline.
- 2026-08-15toolClaudeshipAnthropic's assistant — the practitioner's favorite for prose and long documents.
- 2026-08-15toolClaude CodeshipAnthropic's terminal agent that edits, runs and tests code in your repo.
- 2026-08-15toolSourcegraph CodyskipAI coding assistant with deep code-search context from Sourcegraph.
04 /WHY
Why open source?
02
Receipts, not marketing
Every verdict links to the sources it rests on. Limitations are named on the page, next to the praise, in the same font size.
03
Edit anything via PR
Disagree with a verdict? The catalog is a git repository. Fork it, change it, argue in the open — maintainers answer with data.
05 /CONTRIBUTE
The sky is incomplete without your entry.
Add a tool, correct a verdict, map a stack. One YAML file and a pull request — that is the whole submission process.