Troubleshooting home for /dex-doctor and /feedback (rebased remainder) - #432
Merged
Conversation
What this pull request touches
Gates that will judge this change
Based on 7 changed files. |
…tor's onboarding moment Remaining scope of the feedback-loop-launch-awareness card (PR #431 covered onboarding /feedback intro and post-update announcements): - Canonical troubleshooting passage in Dex_System_Guide.md (Vault Health section): /dex-doctor as first stop when something seems broken, /feedback when it's a genuine Dex bug — with the privacy promise spelled out. - Brief pointers: README setup-troubleshooting block, Dex_Technical_Guide's Testing Doctor section, and dex-doctor SKILL.md Related Commands. - Doctor's onboarding moment: the completion message's 'if anything ever feels broken' beat now introduces /dex-doctor before /feedback, and the getting-started completion list gains a /dex-doctor bullet. - docs/Dex_System copies kept byte-identical to 06-Resources sources. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The original commit was written on Aug 8. Three things have shipped since that change what this copy should say: - Onboarding now has its own feedback + Doctor beat (#454), so the onboarding hunk from the original commit is dropped entirely rather than rebased — it edited a line that no longer exists in that form. - Casual wording now routes to the feedback packager (#455, v1.94.0), so the guides no longer tell people to run a command or say a magic phrase. They say what is now true: describe it however you like. - The help pages for both surfaces are live and verified, so the passages link them instead of describing them. Also tightened two claims: Doctor "repairs what it can on its own without touching your notes" rather than "fixes what's safe" (safe is a word this repo deliberately avoids in user copy about what a tool may touch), and reports "by default wait for your yes" rather than an unqualified promise that everything is always shown, which is not true for users who have explicitly opted into auto-send. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
davekilleen
force-pushed
the
feat/feedback-awareness-remainder
branch
from
August 12, 2026 00:22
73d43db to
722b210
Compare
davekilleen
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Rebased onto current
mainand reconciled against everything that shipped since it was opened on Aug 8. Was conflicting; now clean. Recommendation: merge the remainder (details below), but holding per the Front Desk.What changed in this reconciliation
This PR predated three things that have since shipped, so it has been split: the superseded half is dropped, the still-valuable half is kept and brought up to date.
Dropped as superseded (1 of 6 changes)
.claude/flows/onboarding.md— add/dex-doctorto the completion-message beatKept — genuinely not done anywhere else (5 of 6)
Verified against current
mainbefore keeping:Dex_System_Guide.md,README.mdandDex_Technical_Guide.mdcontain zero mentions of/feedbacktoday. Workstream 4 of thefeedback-loop-launch-awarenesscard is real and still open.06-Resources/Dex_System/Dex_System_Guide.md— the canonical "Troubleshooting: When Something Seems Broken" passage.README.md— pointer in the setup-troubleshooting block, for "something broke after setup".06-Resources/Dex_System/Dex_Technical_Guide.md— pointer from the Doctor testing section..claude/skills/dex-doctor/SKILL.md—/feedbackin Related Commands (the skill body already invokes it twice; the list didn't mention it)..claude/skills/getting-started/SKILL.md— a/dex-doctordiscovery bullet. Its/feedbackbullet was improved separately since Aug 8, so the merge keeps the newer bullet rather than reverting it.docs/Dex_System/mirrors kept byte-identical, astest_docs_bridge.pyrequires.Brought up to date (new commit)
The Aug 8 copy is factually behind
v1.94.0:/feedbackas the explicit alternative.help/feedback.html,help/updating-troubleshooting.html#health-dex-doctor) instead of describing them. Both verified 200 with anchors present./feedback, and that setup problems route back to Doctor.Verified
test_docs_bridge.py(mirror equality),test_skill_integrity.py,test_instruction_honesty.py,test_feedback_trigger_routing.pyNo changelog entry: guide and README copy only, no behaviour change. The behaviour it documents already shipped in 1.93.1–1.93.3 / v1.94.0.
🤖 Generated with Claude Code