Skip to content

feat(optimization): run knowledge searches through complete methods#76

Merged
drewstone merged 1 commit into
mainfrom
feat/official-optimizer-knowledge
Jul 24, 2026
Merged

feat(optimization): run knowledge searches through complete methods#76
drewstone merged 1 commit into
mainfrom
feat/official-optimizer-knowledge

Conversation

@drewstone

Copy link
Copy Markdown
Contributor

Summary

  • run retrieval, full RAG, KB policy, and memory configuration searches through one complete OptimizationMethod
  • retain only bounded exhaustive enumeration for small finite retrieval grids
  • preserve independent train, selection, and final data, canonical candidate identity, resumable artifacts, parallel branches, cost limits, and compare-and-set activation
  • remove knowledge coupling to proposer and local optimizer APIs
  • update concise entry-point docs and the published agent-eval dependency

Verification

  • pnpm lint: 176 files clean
  • pnpm typecheck: pass
  • focused optimization tests: 34 passed
  • pnpm test: 440 passed, 10 skipped
  • pnpm build: pass, including declarations
  • pnpm verify:package: clean install, 5 imports, skill, CLI version, and re-pack
  • deleted optimizer API scan: 0 references
  • git diff --check: pass
  • merge-tree against current origin/main: clean

@tangletools tangletools left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

✅ Auto-approved drewstone PR — 68c4de17

This PR was opened by the trusted drewstone account.
The full PR reviewer audit still runs separately and will publish findings if it detects issues.

tangletools · auto-approval · reason: drewstone_author · 2026-07-24T01:32:50Z

@drewstone
drewstone merged commit 32e1c1b into main Jul 24, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants