Skip to content

fix(opencode): keep the one-shot review out of the mode flag - #1047

Closed
tangsong404 wants to merge 1 commit into
DietrichGebert:mainfrom
tangsong500:fix/opencode-review-latch
Closed

tangsong404 wants to merge 1 commit into
DietrichGebert:mainfrom
tangsong500:fix/opencode-review-latch

Conversation

@tangsong404

Copy link
Copy Markdown

/ponytail review was validated with normalizePersistedMode, so the argument was written to the long-lived .ponytail-active flag. Every later session then injected only the one-line level: review stub instead of the ruleset, until the user switched back to a real level. Review is a one-shot skill, never a persistent level (#377, #736), and every other host already rejects it as a command argument.

Validate the command argument with normalizeMode (runtime levels only). Reading is unchanged: a stored review flag is still honored.

`/ponytail review` was validated with `normalizePersistedMode`, so the
argument was written to the long-lived `.ponytail-active` flag. Every
later session then injected only the one-line `level: review` stub
instead of the ruleset, until the user switched back to a real level.
Review is a one-shot skill, never a persistent level (DietrichGebert#377, DietrichGebert#736), and
every other host already rejects it as a command argument.

Validate the command argument with `normalizeMode` (runtime levels only).
Reading is unchanged: a stored `review` flag is still honored.
@DietrichGebert

Copy link
Copy Markdown
Owner

Thanks! #1103 fixes the same review problem and also resets a review that was already saved, so I'm going with that one.

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