Projects

Forkable automations with an outcome, proof outputs, replay commands, permission scopes, and agent handoff.

3 public blueprints

Featured blueprint

GitHub Projects for AI Engineers

Outcome: Turn public repository observations and ecosystem references into a ranked repository radar and a navigable map of the wider production AI stack.
Wow moment: One HTML or JSON route now exposes 190 ranked repositories plus 98 official tool links across 13 capability lanes and nine production-depth levels.

candidateB2locally-reproduced192 sourcesRelated shift

Time to first result: One local content pass plus a GitHub metric refresh
Estimated cost: Free, except the curation time and GitHub metric refresh
Code level: AI engineer, developer advocate, or technical editor
Write actions: Public site content only after owner-approved deployment
Proof outputs: 6 fixtures, 5 replay commands
Bottleneck: GitHub metrics drift quickly; every refresh needs a current GitHub read and an AI-engineering relevance pass.
Last verified: Sep 13, 2026

Reduce cost per completed coding task without relying on every prompt to say 'read less'.

Turn an implementation idea into a bounded coding task with checks and reviewable evidence.

Status: candidate
Buildability: B2 engineer-repeatable
Time: 1-2 hours in a prepared repository
Cost: Low to medium, depending on model and test loop length
Write actions: Local file edits only; commits and deploys remain explicit approvals
Proof: locally-reproduced

Every blueprint follows the same five-part contract.

Agent handoff

Blueprint Markdown/projects/[slug].md

Build brief/projects/[slug]/build-brief.md

Acceptance checklist/projects/[slug]/acceptance.md

Proof contractproof.fixture_outputs[]

Project contract/projects/[slug].json