Repository navigation
how to prevent default triggering the openspe propose commands even though the repo doesn't initialized the openspec on user promt like explore and propose #1645
Description
Activity
global skills need a way to scope themselves to initialized repos only
- added a commit that references this issue
on Sep 9, 2026 - added a commit that references this issue
on Sep 16, 2026 Fixed on main by #1787, released in v1.13.1. Every workflow now checks for an OpenSpec root before writing anything. If the agent picked the skill itself in a repo with no
openspec/, it stops using OpenSpec and answers normally. Skill descriptions now name OpenSpec too, so hosts are less likely to pick them for unrelated requests. The PR said "Part of" while it was in review, so this didn't auto-close. Thanks for the report!
I see an issues if the user installs the openspec and then prompt AI to explore and propose on any repo, which is by default triggering the openspe propose commands even though the repo doesn't initialized the openspec. how to prevent openspec skill trigger by default even the repo doesn't have the openspec configuration. Is there any solution or guardrails already exist ? I currently using the openspec v1.7.0. My proposal is , if user trying the prompts like explore or propose then skill can first check on that repo config.yaml exist or not, and if not exist it can go through the normal general propose not the openspec.