Owner decision batch 2026-08-31: 7 decisions from the triage wave, one reply resolves all

[hyperpolymath]

Per the 08-27 ruling ("batch owner decisions"), every decision the 2026-08-31 triage wave surfaced, in one place. Each line links the issue holding the full evidence and costed options. Answering here (e.g. "662b, 664a, 698c, 331 delete/~30, 446 yes, 658 publish") is enough — I'll execute and close the loop on each issue.

Decisions needed

1. wordpress-tools PR #57 deleted 108 source files and added nothing — deletion scored as migration #662 — wordpress-tools repo: (a) retire/archive, (b) restore CI to green, or (c) close as-is and mark dormant. (Also carries new Estate drift remediation — 2026-05-16 sweep findings #66 .git_corrupted/ 246 files and Eliminate npm lockfiles estate-wide (npm-avoidant) #67 15 orphaned .res.js — disposition follows the same choice.)
2. 43% of estate .affine files are 200-byte Harvard Engine stubs — migration campaigns count deletions as ports #664 — 63 .affine stub files (83% re-confirmed byte-for-byte): (a) declare debt + tracking issue per repo, (b) restore from .res sources, or (c) port properly under the ReScript enclave ruling.
3. Banned languages appear in dev-environment definitions (guix.scm ×14, empty-linter mise.toml) #698 — banned languages in dev-env configs (~260+ repos): (a) exempt dev-env files from the ban, (b) full sweep, or (c) deno-pins-only sweep (59 files — executable today under the standing Deno→Bun ruling; say the word).
4. [campaign] Estate boj-build.yml sweep — repair or retire (~30 repos with malformed JSON + dead .local host + http://) #331 — AGENTIC.md fleet: two confirms: delete vs replace-with-guarded-template, and scope ~30 (worst offenders) vs ~252 (all).
5. Proposal: an honest, adoptable readiness/insight badge framework — grade our own work, and challenge the evaluators #446 — badge policy: pilot-repo confirm (boj-server). First step (Glama badge removal) is correct under all options and proceeds under the 08-27 pre-authorization.
6. Deno→Bun: all 30 deno.json locations assessed — 18 blocked on npm packages that do not exist #658 blocker-1 (standing since 08-28): affinescript package — publish to npm, switch consumers to git-path, or restore the vendored copy.
7. proven: postulate cure is NOT on origin/main — 42 .idr files/360 sites still unparseable there; the full cure exists only as ~77 UNCOMMITTED working-tree files on a local branch #705 — proven's tree: the 72 whole-file .ts deletions (bindings/deno + bindings/assemblyscript) uncommitted on disk — (a) bless as a banned-language purge and land as an audited commit, or (b) treat as sweep damage and discard. The .idr cure is already safe on fix/idris2-totality-and-elaboration-cures (d2d57a0d) either way; that branch also needs a merge decision.

Not decisions (FYI only)

• lockfile drift: 5 repo(s) as of 2026-08-25 #639 closed (drift reports anonymized repos as _w — permanently unactionable; superseded by Lockfile Drift Detect anonymizes repos as _w, counts banners as entries, treats script errors as drift — and has only ever run once #708).
• New issues from the wave: install-tools.sh regenerated estate-wide WITHOUT the #678 cure — fake exit-0 gate re-propagated to 245/248 copies (fix the template, not the copies) #702 (p1), Duplicate checkout trees double-count every estate census — repos/ vs hyper-repos/ overlap on 213 names, _SET containers hold 412 nested repos #703, examples/web-project-deno.json template ships phantom npm:affinescript@^12.1.0, replicated across ≥12 repos — the purge phantoms are still propagating via scaffold #704, proven: postulate cure is NOT on origin/main — 42 .idr files/360 sites still unparseable there; the full cure exists only as ~77 UNCOMMITTED working-tree files on a local branch #705 (p1), mise.toml pins banned toolchains estate-wide — python in 543/573 files (249 repos); 59 files still pin deno despite the Deno→Bun ruling #706, Phantom upload-artifact pin ea165f8d65b6db9a… (commit never existed, 422) live in 11 repos — spliced mutant of the healthy v4 SHA #707 (p1), Lockfile Drift Detect anonymizes repos as _w, counts banners as entries, treats script errors as drift — and has only ever run once #708, wordpress-tools#66/Eliminate npm lockfiles estate-wide (npm-avoidant) #67.
• Pre-authorized lane proceeds in order Proposal: an honest, adoptable readiness/insight badge framework — grade our own work, and challenge the evaluators #446 (first step) → [#92 Class D] abbreviation / acronym-boundary drift across DB + cloud cartridges #156 → audit: dotfile drift across estate (2026-06-02 sample) — .editorconfig / .gitignore / .gitattributes #343 phase 1 without further asks.