Context
`/track` (`tools/Test-DesignDrift.ps1`) compares every slice in `design/30-slices.md` against the tracker and reports a slice with no matching issue as `[NoIssue]`. A 2026-08-31 run found 48 such slices, all already `### [x]` / Complete:
- W0-W40 (everything before `## Outstanding`, i.e. the `## Done — Specification and Scaffold`, `## Core`, `## The Story-Graph Kind`, `## Content, Clients, Proof` and `## Post-MVP — Depth` sections)
- W61, W63, W64, W65
- W90, W91, W92
None of them have criteria drift — every `Done when` id that exists in the doc for a slice that does have an issue matches exactly. The only finding is the missing issue itself.
This is the same shape #214 already resolved for W41-W49/W50/W51/W66: work that shipped before this repository's issue-per-slice convention existed. That one was closed out by fixing the heading format and opening retroactive issues for the units it covered. This one is broader (48 units, some going back to the very first slices) and the ask here is different: exempt them by design rather than retroactively ticket them, since none of this work will be revisited and 48 closed issues for it would be pure tracker noise.
Ask
Record, here and in this repository's working conventions if it needs restating there, that the slices listed above are exempt from `/track`'s issue sync — their `[NoIssue]` findings are expected and not actionable drift. A future `/track` run should recognise this issue as the standing exemption rather than proposing to open 48 issues again.
Left open rather than closed: nothing about `design/30-slices.md` or `tools/Test-DesignDrift.ps1` was changed to encode the exemption mechanically, so the same 48 `[NoIssue]` findings will keep appearing on every future drift check. That's accepted noise, not a defect this issue fixes — closing it would make the exemption harder to find the next time someone re-derives this reasoning from scratch.
Origin
Raised while running `/track` on 2026-08-31, cross-referencing `design/30-slices.md` against `gh issue list --state all`.
Context
`/track` (`tools/Test-DesignDrift.ps1`) compares every slice in `design/30-slices.md` against the tracker and reports a slice with no matching issue as `[NoIssue]`. A 2026-08-31 run found 48 such slices, all already `### [x]` / Complete:
None of them have criteria drift — every `Done when` id that exists in the doc for a slice that does have an issue matches exactly. The only finding is the missing issue itself.
This is the same shape #214 already resolved for W41-W49/W50/W51/W66: work that shipped before this repository's issue-per-slice convention existed. That one was closed out by fixing the heading format and opening retroactive issues for the units it covered. This one is broader (48 units, some going back to the very first slices) and the ask here is different: exempt them by design rather than retroactively ticket them, since none of this work will be revisited and 48 closed issues for it would be pure tracker noise.
Ask
Record, here and in this repository's working conventions if it needs restating there, that the slices listed above are exempt from `/track`'s issue sync — their `[NoIssue]` findings are expected and not actionable drift. A future `/track` run should recognise this issue as the standing exemption rather than proposing to open 48 issues again.
Left open rather than closed: nothing about `design/30-slices.md` or `tools/Test-DesignDrift.ps1` was changed to encode the exemption mechanically, so the same 48 `[NoIssue]` findings will keep appearing on every future drift check. That's accepted noise, not a defect this issue fixes — closing it would make the exemption harder to find the next time someone re-derives this reasoning from scratch.
Origin
Raised while running `/track` on 2026-08-31, cross-referencing `design/30-slices.md` against `gh issue list --state all`.