auto-improve

GAN-style self-improvement loop for any text artifact: mutate, grade with a SEPARATE model, keep only verified wins (pairwise-judged), revert the rest. The git history is the improvement log.

crimeacs/auto-improve

View on GitHub ↗

GAN-style self-improvement loop for any text artifact: mutate, grade with a SEPARATE model, keep only verified wins (pairwise-judged), revert the rest. The git history is the improvement log.

auto-improve grew +2.9% in the last 30 days, ranks 865th of 1,105 in Prompting & Scaffolding, actively maintained, last commit 4 days ago.

Star trajectory

82 days tracked
2026-08-31141 stars today2026-11-22
Momentum

+2.9% in 30d, above the Prompting & Scaffolding median of +1.5%.

Maintenance

Last commit 4 days ago, fresher than the median of 35d.

How it compares in Prompting & Scaffolding

auto-improve
141
stars
+2.9% 30d
commit 4 days ago
Prompting & Scaffolding median
319
stars
+1.5% 30d
commit ~35d ago
Leader · ponytail
196.6k
stars
+13.0% 30d
commit today

Ranked 865th of 1,105 by stars.

Ranked by shared tags, then keywords, then stars
  1. 🔁 Improve your agent harness autonomously overnight

    +7 (+2.2%) in 30d · Committed 208d ago · Python · MIT

    Shares Evaluation & Guardrails, Prompting & Scaffolding

  2. Alignment-research scaffold (autoresearch-style) for LLM guardrails: search over a single policy.md surface

    -1 (-0.8%) in 30d · Committed 52d ago · Python · Apache-2.0

    Shares Evaluation & Guardrails, Prompting & Scaffolding

  3. Bounded, evidence-driven SKILL.md evolution under frozen evaluation and safety contracts.

    -1 (-1.1%) in 30d · Committed 68d ago · TypeScript · MIT

    Shares Evaluation & Guardrails, Prompting & Scaffolding

  4. 🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓

    +59 (+0.9%) in 30d · Committed 4d ago · TypeScript · Apache-2.0

    Shares Evaluation & Guardrails, Prompting & Scaffolding

  5. Open-Source Agent Observability & Evals for AI Platform teams. Trace, test, route and govern every agent &…

    +63 (+1.3%) in 30d · Committed today · TypeScript · Apache-2.0

    Shares Evaluation & Guardrails, Prompting & Scaffolding

  6. An open-source visual programming environment for battle-testing prompts to LLMs.

    +6 (+0.2%) in 30d · Committed 23d ago · TypeScript · MIT

    Shares Evaluation & Guardrails, Prompting & Scaffolding

  7. UpTrain is an open-source unified platform to evaluate and improve Generative AI applications. We provide…

    -1 (-0.0%) in 30d · Committed 825d ago · Python · Apache-2.0

    Shares Evaluation & Guardrails, Prompting & Scaffolding

  8. The Fable Workflow: how Claude Fable 5 worked, distilled into skills any model can run, with the eval that…

    +19 (+0.8%) in 30d · Committed 50d ago · Python · MIT

    Shares Prompting & Scaffolding, Evaluation & Guardrails

  9. Turn any goal into a short prompt that makes your agent set a real quality bar, run builder and critic pairs…

    +119 (+10%) in 30d · Committed 108d ago · CC-BY-4.0

    Shares Prompting & Scaffolding, Evaluation & Guardrails

  10. The Continuous-Improvement Stack for Agents. Our environment data and evals power agent improvement and…

    +4 (+0.4%) in 30d · Committed 2d ago · Python · Apache-2.0

    Shares Evaluation & Guardrails, Prompting & Scaffolding

Language
Python
License
MIT License
Contributors
3
Forks
16
Watchers
0
Open issues
0
Releases
1
Created
2026-06-01
Stars this week
+1

You've seen where auto-improve stands today.

Get the moves before everyone else: the 5 Prompting & Scaffolding repos climbing fastest, every Tuesday.

Free weekly digest. Double opt-in, unsubscribe anytime.

Maintain auto-improve?

Claim this page to add your own call-to-action.

Claim this page