Skip to content

Commit f4aba32

Browse files
TheWayWithinclaude
andcommitted
docs: v0.3.0 tagged ahead of human gates by owner decision — gates now post-release validation
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
1 parent c8e0ded commit f4aba32

2 files changed

Lines changed: 22 additions & 24 deletions

File tree

CLAUDE.md

Lines changed: 7 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -60,9 +60,13 @@ All specs live in `ideation/` (moved from repo root, 17 Jul 2026):
6060
build (checksum-pinned) on Ubuntu for the batchpass run.
6161
- v0.2 history: commits d9b94a9 + b4725df (schema v2, two-tier validation,
6262
setup/review orchestration).
63-
- Awaiting Jamie: the two RELEASE-CHECKLIST.md gates (UAT plan:
64-
docs/UAT-PLAN.md; tasks T-159..T-166). Consolidation done 17 Jul 2026:
65-
`digital-estate` (Arm A) archived on GitHub with a pointer README.
63+
- **v0.3.0 TAGGED + RELEASED 18 Jul 2026** ahead of the two human gates,
64+
by deliberate owner decision (live end-to-end testing; zero traffic).
65+
Release notes state machine-verified vs pending honestly. The gates
66+
(UAT plan: docs/UAT-PLAN.md; tasks T-159..T-166) remain open as
67+
post-release validation blocking the "safe to recommend" claim;
68+
findings → v0.3.1. Consolidation done 17 Jul 2026: `digital-estate`
69+
(Arm A) archived on GitHub with a pointer README.
6670
- Session learnings worth keeping: `expect -` treats the first argument as a
6771
script file, so heredoc expect scripts must take filenames via env vars,
6872
not argv; `tr < /dev/urandom | head` dies with SIGPIPE (141) under

RELEASE-CHECKLIST.md

Lines changed: 15 additions & 21 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,14 @@
11
# v0.3 "Executor Release" — release checklist
22

3-
Everything below the line is done, tested, and green on main. **Two
4-
gates remain, and both need another human.** Do not tag `v0.3.0` until
5-
both pass; fold every observed stumble back into the docs first.
3+
**Status update 2026-07-18:** `v0.3.0` was tagged and released AHEAD of
4+
the two human gates, by deliberate owner decision — pre-launch, zero
5+
traffic, and live end-to-end testing of the full pipeline (repo → tag →
6+
site download) is easier against the real release. The release notes
7+
state plainly which verification is machine-proven and which is
8+
pending. **The two gates below are unchanged in substance and remain
9+
open as post-release validation** — they now block the "safe to
10+
recommend to real families" claim and v0.3.1, not the tag. Findings
11+
fold into v0.3.1.
612

713
## Gate 1 — the Windows dry run (THE release gate)
814

@@ -41,25 +47,13 @@ re-run `scripts/make-guide.sh` and re-print so the guide carries the
4147
- [ ] Fire drill passed with real printed shares
4248
- [ ] Date + tester recorded and visible on the re-printed guide
4349

44-
## Then tag
50+
## Tagged (2026-07-18)
4551

46-
```
47-
git tag -a v0.3.0 -m "Executor Release — schema 3; recovery tested on Windows + paper"
48-
git push origin v0.3.0
49-
```
50-
51-
Create the GitHub release with:
52-
53-
- **Version:** v0.3.0 · **Schema:** format 3 (format 2 accepted with
54-
migrate warnings for this version)
55-
- **Platforms tested:** macOS (CI + local), Ubuntu (CI), Windows 10/11
56-
(manual dry run — name the date)
57-
- **Tool versions tested:** age 1.3.1 (batchpass) and age 1.1.x
58-
(stock interactive, Ubuntu CI); ssss 0.5
59-
- **Checksums:** attach `shasum -a 256` of the tagged tarball
60-
(`git archive --format=tar.gz -o executor-file-v0.3.0.tar.gz v0.3.0`)
61-
- **Migration notes:** copy the "MIGRATING FROM FORMAT 2" block from
62-
`schema/estate.schema.yaml`
52+
`v0.3.0` is live: https://github.com/TheWayWithin/executor-file/releases/tag/v0.3.0
53+
— tarball + `.sha256` attached, release notes honest about the pending
54+
gates. When both gates pass: fold findings into the docs, release
55+
v0.3.1 with "Windows 10/11 (manual dry run — date)" added to platforms
56+
tested, and update the site.
6357

6458
---
6559

0 commit comments

Comments
 (0)