Boolean metrics: fix formatting and exclude from fitness; README badg…

…es (#473)

Booleans are flags, not scores. Renders bool metrics as True/False, skips them in improvement diffs, and excludes them from safe_numeric_average/get_fitness_score so a timed-out program ({error: 0.0, timeout: True}) scores 0.0 instead of 0.5. Removes dead _format_metrics/_format_improvement helpers from controller.py. README: pepy downloads badge (grey/green), stars badge dropped. Bumps version to 0.3.2.

run_evolution checkpoint params; propagate llm.provider (#472); backg…

…round_blur example (#457)

Plumbs target_score/checkpoint_path through run_evolution(); fixes llm.provider not propagating to per-model configs (supersedes #472); adds the background_blur example (hot function behind a hard quality gate). Bumps version to 0.3.1.

Fix MAP-Elites eviction (#454) & sampling (#452) bugs; dhara CI + Fra…

…me SAST (#462)

Stacks the initial-program-artifacts fix with: MAP-Elites elite protection and zombie removal (#454); program-sampling cleanup incl. removing dead iteration.py (#452); lambda-evaluator SyntaxError fix; swap integration model to public codelion/dhara-250m with bounded max_tokens; full integration suite in CI; Frame SAST workflow. Bumps version to 0.3.0.

Fix bugs (#442)

* Fix library API evaluators broken with process-based parallelism

The callable evaluators in evolve_function and evolve_algorithm used
closure-based functions stored in globals(), which don't survive across
process boundaries. Since the switch to process-based parallelism
(c2f668a), subprocess workers cannot access the parent process memory,
causing "module has no attribute '_openevolve_evaluator_*'" errors.

Fix by serializing evaluators as self-contained code strings instead of
closures. Also adds combined_score to returned metrics to prevent the
misleading score averaging warning.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* Bump version to 0.2.27 and add subprocess evaluator tests

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* Update api.py

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Make max snapshot artifacts limit configurable (#386)

* Make max snapshot artifacts limit configurable

Add `database.max_snapshot_artifacts` config option to control how many
program artifacts are included in worker process snapshots. Default
remains 100 for backward compatibility.

- Set to a higher number to include more artifacts in prompts
- Set to `null` (None) for unlimited artifacts (use with caution for
  large populations as this can significantly increase memory usage)

Note: This limit only affects artifacts passed to worker processes,
not the total artifacts stored. All program code is always available
regardless of this setting.

Closes #383

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* Add tests for recent features

Add comprehensive tests for recently merged PRs:

- test_llm_config_optional_params.py: Tests for optional temperature/top_p
  parameters (PR #385 - Anthropic model compatibility)

- test_snapshot_artifacts_limit.py: Tests for configurable max_snapshot_artifacts
  (PR #386)

- test_visualization_sanitization.py: Tests for -inf/+inf/NaN sanitization
  in visualization (PR #384)

- test_early_stopping_config.py: Tests for event-based early stopping
  configuration (PR #375)

- test_changes_description.py: Tests for large codebase support via
  changes description (PR #376)

Total tests increased from 264 to 326.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* Add integration tests for example validation

Add comprehensive integration tests that verify:
- Example config files load correctly
- Initial programs have EVOLVE-BLOCK markers
- Evaluators exist and have required functions
- Evaluators can run on initial programs
- Cascade evaluation functions are detected
- Database stores and retrieves programs correctly
- Program evolution tracking works

Tests cover function_minimization, circle_packing, and signal_processing
examples, plus general structure validation for all examples.

Total tests: 346 (was 326)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

Merge pull request #363 from FunMelon/refactor-template-prompt

refactor(prompt): Replace hardcoded strings with template fragments

Merge pull request #358 from algorithmicsuperintelligence/fix-new-rel…

…ease

bump version for new release

Merge pull request #345 from algorithmicsuperintelligence/fix-better-…

…prog-calculation

Update controller.py

Merge pull request #329 from algorithmicsuperintelligence/fix-key-emb…

…edding-model

Fix key embedding model

Merge pull request #328 from algorithmicsuperintelligence/fix-embeddi…

…ng-async-error

Fix embedding async error