Skip to content

Feature/memrok openclaw compatibility fixes#3

Open
christophermallory wants to merge 3 commits intomemrok-com:mainfrom
christophermallory:feature/memrok-openclaw-compatibility-fixes
Open

Feature/memrok openclaw compatibility fixes#3
christophermallory wants to merge 3 commits intomemrok-com:mainfrom
christophermallory:feature/memrok-openclaw-compatibility-fixes

Conversation

@christophermallory
Copy link
Copy Markdown

Fix for issue #2.

  • Switches to node:sqlite for compatibility with Openclaw plugin installer.
  • Removes hard-coded file paths and uses Openclaw agent directories for scanning memories.
  • Allows the use of Openclaw providers and models or overriding via config.
  • Adds Openclaw slash commands for manual reindexing.
  • Updates readme.md for openclaw-plugin to reflect changes.

Copy link
Copy Markdown
Contributor

@MichaelSchmidle MichaelSchmidle left a comment

Choose a reason for hiding this comment

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

Thanks, this is a genuinely useful contribution and the overall direction makes sense, especially the node:sqlite switch and the OpenClaw compatibility cleanup.

I can’t merge it as-is yet because the branch currently fails to parse/compile, with blocking issues in packages/store/src/store.ts and packages/store/src/schema.ts.

Once that’s cleaned up, I’d be happy to take another look.

@MichaelSchmidle
Copy link
Copy Markdown
Contributor

MichaelSchmidle commented Apr 15, 2026

Thanks again for this, and sorry the branch drift got worse while we were moving quickly on main.

I’ve now routed issue #2 into our v0.7.0 milestone, and I still think the core direction in your PR is valuable, especially around OpenClaw compatibility.

At this point I don’t think we should try to merge the whole PR as one large blob anymore, because has changed a lot in the plugin path since you opened it and the branch is now conflicting in several of the same files.

If you’re still up for it, the most helpful next step would be to rebase and narrow this PR to the compatibility core for #2:

  • installer/runtime compatibility
  • path discovery / indexing correctness
  • provider/model inheritance

The more optional operator surface (manual commands, broader polish, etc.) can be separate if needed.

If you’d rather not keep chasing a moving target, that’s also totally understandable. Either way, thank you, this issue helped surface a real gap in the product and shaped how we’re planning the next arc.

@christophermallory
Copy link
Copy Markdown
Author

I don't mind contributing. I'll free up in a few days and take a look.

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