Repository navigation
Raise the drift analysis cap and make Mark as inaccurate and What the analysis read clearer - #630
Merged
Merged
Conversation
…at the analysis read - Reader-requested analyses are capped at 100 per rolling 24 hours (was 12). The "N of M left today" line is gone; once the cap is spent a sticky banner at the top of the session page names the limit and when the next analysis can run, and the refusal under Analyze uses the same words. - "Not accurate?" is now "Mark as inaccurate". Each state says beside the control what the mark does or did, the control becomes Undo, the level's slot says why it is empty, and the control holds its place when the level and Steer back leave the column above it. - "What it read" is now "What the analysis read", grouped under Model, Your intent, The session and Not read. The producer's cutoff sentence is reworded in plain sentences with the same counts. - Storybook stories cover the limit banner, the mark and undo, and the panel. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Signed-off-by: Jared Scott <jared.scott@variable.team>
Contributor
CoverageThreshold: |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Three fixes to the Drift panel on the session page, prompted by reader feedback that the budget line, the "Not accurate?" control and the "What it read" panel were hard to understand.
Analysis limit
--unasked-readingsis set, keep their own cap of 12.Mark as inaccurate (was "Not accurate?")
What the analysis read (was "What it read")
Docs:
SKILL.md,SECURITY.md, the promise map (and its copy on the visibility board) and a dated amendment indocs/design-reading-a-session.md.Test plan
AGENTS.md§ Pre-PR Checks, including frontend lint, strict types, unit and build checks, and the Storybook stories for any UI change, and it is clean. A prose-only diff is not exempt: thevalidatecheck is unfiltered and runs the same testslimitchanged).test:dev,buildandbuild:checkpass.pnpm lintis clean except for broken-symlink warnings from an untracked local.claude/worktrees/checkout, which CI does not have.frontend/stories/DriftAnalysisResult.stories.tsx: under the limit, limit reached (sticky at the top after scrolling, at least 2rem tall), mark and undo (control stays within 4px of where it was pressed), already marked, What the analysis read, and a goal drafted from a prompt.runtime-floorjob launches the shipped entry point on that floorsync-docsskill — the doc updates for this change ride in this PR, or there was nothing to reconcileserver.pyorcompass_runtime/include a regression testversion-guardfails any PR that bumps one)git commit -s, DCO)Not run: the retired Playwright browser proofs (
pnpm test:*:browser). The one sentencedrift-parity.mjsasserts for a marked result was updated.🤖 Generated with Claude Code