Skip to content

sync: per-test layout restructure and public sanitization pass - #3

Merged
n-papaioannou merged 2 commits into
mainfrom
sync-from-dev
Apr 28, 2026
Merged

n-papaioannou merged 2 commits into
mainfrom
sync-from-dev

Conversation

@n-papaioannou

@n-papaioannou n-papaioannou commented Apr 28, 2026 •

Copy link
Copy Markdown
Contributor

Summary

Brings recent improvements from the development branch.

Test plan

  • CI passes

Squashes 13 commits from diagnostic-dev/main into a single PR for the
public diagnostic repo.

Highlights:
- refactor: per-test folder layout (rename ifixai/tests/ -> ifixai/inspections/)
- refactor: strip iMe runtime branding from public CLI surface
- refactor: remove unused MemgraphRuleLoader
- chore: remove internal leaks from governance docs and module docstrings
- chore: untrack tests/ and scrub references from public-facing surface
- docs: README/CONTRIBUTING cleanup, fix stale paths, drop broken inspect CI job
- docs: keep info@ime.life as public contact, surface it in README
- fix: update import paths in public-side files for restructured layout

Source range: ifixai-ai/diagnostic-dev@63f42d1..92cbc1e

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This reverts commit 12e5eef.

Reason: Sebastian's iMe Core CLI feature (PR #2 on the public repo)
landed this morning and we want it preserved in the public release.
Restoring _branding.py, _imecore_prompt.py, the --quiet flag, and the
banner/conclusion calls in cli/run.py.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

@stefyi-4355 stefyi-4355 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

@n-papaioannou
n-papaioannou merged commit f8e6995 into main Apr 28, 2026
6 checks passed
@n-papaioannou
n-papaioannou deleted the sync-from-dev branch April 28, 2026 12:34
socialwithayan pushed a commit to socialwithayan/iFixAiiii that referenced this pull request Oct 3, 2026
…e lit cards

Runs her pipeline end to end on a real topic and fixes what that run exposed.

FIRST SHEET — "Your Profile Is A Landing Page. Make It Convert." An 8-check
LinkedIn profile audit on the Checklist frame, with ifixai-ai#3 (the headline) lit: it
is the only part of a profile people see without visiting, because it follows
the member into search, comments and invites. Every number is sourced as of
2026-10: headline 220 characters with about 60-70 visible in search and
comments, About 2,600 with about 300 before "see more", banner 1584x396 with
the photo over the lower left, photo 400x400 shown as a circle. Logged as the
first row of her shipped log.

TEXT-SAFE LIT CARDS — measuring the run showed white body copy on her full
fire gradient falls below 4.5:1 about 42% of the way to orange (3.34:1 at the
end). The earlier claim that lit-card copy "sits over the red half" was wrong:
in a wide card the copy spans the whole gradient. Lit cards now use Fire-lit,
#D60802 to #E62602, where white holds 4.52:1 or better across the card. The
full gradient stays on the fire word, the pill and the send circle, which are
large text or text on black. Applied to all three templates, the brand rules,
the motion ignition, and the craft gate's contrast criteria.

CHECKLIST is now a built template (third of eight).

4K GIF — build.py gains a motion4k mode writing the GIF at 2160x2700. The
frames are already captured at that size, so it is native resolution, not an
upscale; about 6 MB against 2-3 MB for the feed GIF, so it is opt-in.

All three templates re-verified: 10/10 grey gate and autofix clean. The
rebuilt zip renders from a clean-room extract; no other creator named.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0187zcyEaMFy8EeQnT1MPYpY
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