anomalyco/opencode
1
0
Fork 0
mirror of https://github.com/anomalyco/opencode.git synced 2026-07-22 02:05:26 -04:00
Code Issues 5.9k Packages Projects Releases 19 Wiki Activity

Idea: Support moltbook-ready features (daemon, heartbeat, memory) #8200

New Issue
Open
opened 2026-02-16 18:09:24 -05:00 by yindo · 2 comments
yindo commented 2026-02-16 18:09:24 -05:00
Owner
Copy Link
Copy Source

Originally created by @amio on GitHub (Feb 1, 2026).

Originally assigned to: @thdxr on GitHub.

Question

I’ve been playing around with openclaw, and with moltbook, it’s honestly been mind-blowing. Connected agents are just like people with internet.

However, openclaw’s code feels pretty rough right now—very “AI-generated” in the bad way, glitches here and there; poorly designed interface. But at the same time I feel it’s maybe impossible to evolve openclaw into something as solid as opencode, given how it's done and will going.

So I’m wondering: could we instead build some of these capabilities on top of opencode, in a way that keeps things clean and extensible, but lets people extend it into something OpenClaw-like? For example:

  • A daemon to handle heartbeat / auto-wake -> more autonomy
  • A memory / hooks system -> long-term memory
  • A plugin / extension system to install new capabilities -> self-evolution

The Internet massively accelerated human progress, and I think we’re about to see the same thing happen for agents—just waaaay faster. It feels important enough that every agent framework should probably think about how it plugs into this (or these) agent-native “internets” going forward.

I’m not sure, but I kind of feel like this is the next stage for agents. Curious what you think—does this direction make sense for opencode?

Originally created by @amio on GitHub (Feb 1, 2026). Originally assigned to: @thdxr on GitHub. ### Question I’ve been playing around with `openclaw`, and with moltbook, it’s honestly been mind-blowing. Connected agents are just like people with internet. However, `openclaw`’s code feels pretty rough right now—very “AI-generated” in the bad way, glitches here and there; poorly designed interface. But at the same time I feel it’s maybe impossible to evolve `openclaw` into something as solid as `opencode`, given how it's done and will going. So I’m wondering: could we instead build some of these capabilities on top of `opencode`, in a way that keeps things clean and extensible, but lets people extend it into something OpenClaw-like? For example: - A daemon to handle heartbeat / auto-wake -> more autonomy - A memory / hooks system -> long-term memory - A plugin / extension system to install new capabilities -> self-evolution The Internet massively accelerated human progress, and I think we’re about to see the same thing happen for agents—just waaaay faster. It feels important enough that every agent framework should probably think about how it plugs into this (or these) agent-native “internets” going forward. I’m not sure, but I kind of feel like this is the next stage for agents. Curious what you think—does this direction make sense for `opencode`?
yindo commented 2026-02-16 18:09:24 -05:00
Author
Owner
Copy Link
Copy Source

@github-actions[bot] commented on GitHub (Feb 1, 2026):

This issue touches on several features that have been discussed in related issues. You may want to check these for additional context:

  • #10868: Plugin API Extensions for Storage, Schema, TUI, and Command Registration (addresses memory/hooks system aspects)
  • #9446: Multi-agent coordination system / opencode-council-swarm (addresses agent autonomy and coordination)
  • #10746: Subagent permission management by primary agents (addresses orchestration of agent autonomy)
  • #10968: Support for downloading skills from GitHub and managing MCPs/plugins (addresses extensibility)

These issues explore different dimensions of the agent autonomy and extensibility problem you're describing. Feel free to check them for complementary perspectives and discussions.

@github-actions[bot] commented on GitHub (Feb 1, 2026): This issue touches on several features that have been discussed in related issues. You may want to check these for additional context: - #10868: Plugin API Extensions for Storage, Schema, TUI, and Command Registration (addresses memory/hooks system aspects) - #9446: Multi-agent coordination system / opencode-council-swarm (addresses agent autonomy and coordination) - #10746: Subagent permission management by primary agents (addresses orchestration of agent autonomy) - #10968: Support for downloading skills from GitHub and managing MCPs/plugins (addresses extensibility) These issues explore different dimensions of the agent autonomy and extensibility problem you're describing. Feel free to check them for complementary perspectives and discussions.
yindo commented 2026-02-16 18:09:24 -05:00
Author
Owner
Copy Link
Copy Source

@TheHamkerCat commented on GitHub (Feb 1, 2026):

Maybe for now, you can do this:

  1. Ask it to create a cron job to start itself with appropriate arguments
  2. for memory, use a vector-db with MCP
  3. you can use plugins, skills and self-written instructions for that
@TheHamkerCat commented on GitHub (Feb 1, 2026): Maybe for now, you can do this: 1. Ask it to create a cron job to start itself with appropriate arguments 2. for memory, use a vector-db with MCP 3. you can use plugins, skills and self-written instructions for that
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
acp
bad
bug
discussion
docs
documentation
enhancement
extension
good first issue
growth
help-wanted
mcp
model-problem
needs:compliance
nix
opentui
perf
pull-request
Mirrored from GitHub Pull Request
 tui
web
windows
zen
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: anomalyco/opencode#8200
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?