Rename strategy from AlphaEvolve to OpenEvolve and credit the inspiration

[mrjf]

Summary

Rename the alphaevolve strategy to openevolve and credit the actual inspiration — algorithmicsuperintelligence/openevolve (6K⭐, "Open-source implementation of AlphaEvolve"). AlphaEvolve is the DeepMind paper; OpenEvolve is the open-source implementation our playbook was actually modeled on. The name should reflect what the code is, and the reference links should point contributors to the right place to learn more.

Changes

Directory / file renames

Old path	New path
.autoloop/strategies/alphaevolve/	.autoloop/strategies/openevolve/
.autoloop/strategies/alphaevolve/strategy.md	.autoloop/strategies/openevolve/strategy.md
.autoloop/strategies/alphaevolve/CUSTOMIZE.md	.autoloop/strategies/openevolve/CUSTOMIZE.md
.autoloop/strategies/alphaevolve/prompts/mutation.md	.autoloop/strategies/openevolve/prompts/mutation.md (no rename inside prompts/)
.autoloop/strategies/alphaevolve/prompts/crossover.md	.autoloop/strategies/openevolve/prompts/crossover.md (no rename inside prompts/)
.autoloop/programs/tsb-perf-evolve/strategy/alphaevolve.md	.autoloop/programs/tsb-perf-evolve/strategy/openevolve.md

Use git mv so history is preserved on the moved files.

Text replacements

Throughout all moved files, all prose in .github/workflows/autoloop.md, and every tsb-perf-evolve file:

• AlphaEvolve → OpenEvolve (proper noun references — titles, section headers, narrative prose).
• alphaevolve → openevolve (filenames, path segments, directory names in prose).
• alphaevolve.md → openevolve.md (references to the strategy file itself).

Attribution block

Add this block near the top of both .autoloop/strategies/openevolve/strategy.md and .autoloop/strategies/openevolve/CUSTOMIZE.md:

> **Inspiration.** This strategy is modeled on [OpenEvolve](https://github.com/algorithmicsuperintelligence/openevolve) — an open-source implementation of the evolutionary-code-search approach popularized by DeepMind's AlphaEvolve paper. We've adapted the core ideas (MAP-Elites niching, island model, four operators — exploitation / exploration / crossover / migration) into a playbook the autoloop agent follows at iteration time. Consult the OpenEvolve repo for background on the underlying algorithm and worked examples.

Place it right after the title, before the "This file is the runtime playbook…" sentence.

program.md pointer block

Where tsb-perf-evolve/program.md currently says:

## Evolution Strategy

This program uses the **AlphaEvolve** strategy. On every iteration, read `strategy/alphaevolve.md` and follow it literally …

change to:

## Evolution Strategy

This program uses the **OpenEvolve** strategy (modeled on [openevolve](https://github.com/algorithmicsuperintelligence/openevolve)). On every iteration, read `strategy/openevolve.md` and follow it literally …

And update the Support files list to reference strategy/openevolve.md.

.github/workflows/autoloop.md

The "Strategy discovery" prompt section added by the strategy system doesn't name a strategy, so no change needed there. But audit the file for any example mention of AlphaEvolve and replace.

Backwards compatibility

No autoloop installation outside this repo depends on the strategy name, so no compat shim is needed. The running agent reads whatever file program.md points it at — renaming both the strategy file and the pointer in one PR is safe.

The existing open program tsb-perf-evolve has state on the memory/autoloop branch that may mention "AlphaEvolve" in its iteration history / lessons learned. That's history, not behaviour — leave it as-is, don't rewrite the state file. Future iteration entries will naturally use the new name.

Acceptance

• No file, directory, or markdown text in the repo contains AlphaEvolve / alphaevolve / alphaevolve.md (git-grep for all three, case-insensitive, returns zero hits outside .git/ and historical state files on memory/autoloop).
• Both the strategy playbook and CUSTOMIZE.md carry the attribution block linking to the OpenEvolve repo.
• The tsb-perf-evolve program's program.md pointer block references strategy/openevolve.md and the playbook title matches.
• Next autoloop iteration for tsb-perf-evolve successfully reads strategy/openevolve.md (not 404), no behavioural change.

Related

• Add strategy system to autoloop — ship AlphaEvolve as the first specialized iteration playbook #181 — introduced the strategy; shipped under the (incorrect) AlphaEvolve name.
• tsb-perf-evolve: close the gaps for end-to-end AlphaEvolve loop on Series.sortValues #183 — tsb-perf-evolve implementation details; needs the strategy-file reference updated in lockstep.
• Add Test-Driven as the second autoloop strategy (red/green/refactor playbook) #184 — Test-Driven as a second strategy; adds context about naming conventions — confirm the sibling test-driven name is fine (it is; nothing to rename there).