Problem
The first README flow is currently a successful cellfence init / cellfence check path. That is useful, but it does not immediately demonstrate CellFence's core value: ordinary project checks can pass while an architectural boundary violation is still blocked.
A first-time reader should be able to see, near the top of the README, that CellFence catches a private cross-cell import even when the code still typechecks and tests pass.
Expected behavior
Update the first runnable README demo so it shows this sequence:
- project tests pass
- TypeScript typecheck passes
- CellFence fails with
CELLFENCE_PRIVATE_IMPORT
The demo should make the contrast obvious without requiring a large fixture or prior CellFence knowledge.
Example intent:
npm test
# passes
npm run typecheck
# passes
npx cellfence check
# fails with CELLFENCE_PRIVATE_IMPORT
The exact commands can differ if the demo remains copy-pasteable and accurate.
Scope
Keep this focused on README onboarding.
Do not change CellFence rule behavior.
Do not change the manifest schema.
Do not add a large new example application unless it is clearly needed for the README flow.
Suggested starting points
Likely areas:
README.md
examples/minimal/README.md
examples/minimal/cellfence.manifest.json
The existing README sections to review are:
Try it in sixty seconds
Catch a violation in thirty seconds
Acceptance criteria
- The first README demo clearly shows tests passing, typecheck passing, and CellFence failing.
- The failure shown is
CELLFENCE_PRIVATE_IMPORT.
- The example includes enough files or commands for a reader to reproduce it.
- The README still keeps a short successful init/check path, either before or after the failure demo.
- No product behavior is changed.
- No generated benchmark or research output is added.
Validation
Run:
If the README includes a copy-paste shell demo, manually run that demo in a temporary directory and report the result in the PR.
Problem
The first README flow is currently a successful
cellfence init/cellfence checkpath. That is useful, but it does not immediately demonstrate CellFence's core value: ordinary project checks can pass while an architectural boundary violation is still blocked.A first-time reader should be able to see, near the top of the README, that CellFence catches a private cross-cell import even when the code still typechecks and tests pass.
Expected behavior
Update the first runnable README demo so it shows this sequence:
CELLFENCE_PRIVATE_IMPORTThe demo should make the contrast obvious without requiring a large fixture or prior CellFence knowledge.
Example intent:
The exact commands can differ if the demo remains copy-pasteable and accurate.
Scope
Keep this focused on README onboarding.
Do not change CellFence rule behavior.
Do not change the manifest schema.
Do not add a large new example application unless it is clearly needed for the README flow.
Suggested starting points
Likely areas:
The existing README sections to review are:
Acceptance criteria
CELLFENCE_PRIVATE_IMPORT.Validation
Run:
npm run build npm testIf the README includes a copy-paste shell demo, manually run that demo in a temporary directory and report the result in the PR.