You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hookbridge’s purpose is simple: define hook behavior once, then compile to host-specific plugin outputs (currently Claude Code + Codex), with explicit reporting when behavior cannot be translated 1:1.
I’m sharing implementation feedback from this real-world compiler because I think it can help improve Open Plugin interoperability while the spec is still evolving.
Why this might be useful to the spec
When you actually compile one source model into multiple host formats, you run into practical issues that are not obvious from syntax alone:
Event names may match, but trigger timing/semantics often differ
Some hook types exist on one host and not another
“Equivalent” behavior sometimes requires shims or cannot be represented safely
From our side, this is exactly where authors get hurt, so we expose those gaps directly instead of failing silently.
What I’m exploring
I’m considering adding an Open Plugin compatibility target to Hookbridge in phases:
Emit vendor-neutral .plugin/plugin.json
Support hook-focused Open Plugin packaging paths/conventions
Keep host-specific outputs where needed for behavior fidelity
Avoid claiming full conformance until coverage is genuinely broad
Feedback I’d appreciate from maintainers
Does a hook-focused compatibility profile fit your intended direction?
Would a documented “portable hooks subset” be useful for plugin authors?
Would you accept reference fixtures/test cases from Hookbridge that show:
fully portable cases
partially portable cases
non-portable cases with rationale
Is Discussion-first the preferred contribution path before drafting PRs?
Offer to contribute
If helpful, I can share:
concrete source schema examples
generated Open Plugin + host-specific outputs side by side
a portability matrix that distinguishes syntax compatibility from runtime semantic compatibility
I’m not looking to promote a proprietary product here. Hookbridge is open-source/free, and the goal is to contribute practical interoperability data to make the spec stronger for everyone.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Hi Open Plugin Spec team,
I maintain Hookbridge (https://github.com/REPOZY/Hookbridge), an open-source, free tool focused on plugin hooks.
Hookbridge’s purpose is simple: define hook behavior once, then compile to host-specific plugin outputs (currently Claude Code + Codex), with explicit reporting when behavior cannot be translated 1:1.
I’m sharing implementation feedback from this real-world compiler because I think it can help improve Open Plugin interoperability while the spec is still evolving.
Why this might be useful to the spec
When you actually compile one source model into multiple host formats, you run into practical issues that are not obvious from syntax alone:
From our side, this is exactly where authors get hurt, so we expose those gaps directly instead of failing silently.
What I’m exploring
I’m considering adding an Open Plugin compatibility target to Hookbridge in phases:
.plugin/plugin.jsonFeedback I’d appreciate from maintainers
Offer to contribute
If helpful, I can share:
I’m not looking to promote a proprietary product here. Hookbridge is open-source/free, and the goal is to contribute practical interoperability data to make the spec stronger for everyone.
Thank you for your time and honesty.
REPOZY
All reactions