Refs #1.
`scripts/install-apt.sh` explained the sqv rejection as a malformed MPI
encoding in an Ed25519 key. The registry does not serve one: at
`api/packages/heavy-duty/debian/repository.key`, `gpg --list-packets`
reports `algo 1` (RSA) with a 2048-bit `pkey[0]`. The explanation was
therefore wrong about the only part of itself that was checkable.
The comment now states the measurement and points at #1 for the cause,
rather than naming a mechanism nobody verified. This satisfies #1's
criterion that `git grep -in 25519` return no hits outside test fixtures,
which it now does.
Behaviour is unchanged: this commit edits a comment. The automatic
fallback this file still performs contradicts ruling B and is NOT fixed
here — that work is still owed on !38.
#39 moved the two workflow callers to 0.6.3 but left the doctrine mirror at
0.6.1, so BUILDER.md was missing the three clauses that govern signal-then-ready
(ceremony#330, the undraft order, and ceremony#336). Re-vendored all six
manifest files and moved the pin recorded in .ceremony/README.md.
#39 moved the two workflow callers to 0.6.3 but left the doctrine mirror at
0.6.1, so BUILDER.md was missing the three clauses that govern signal-then-ready
(ceremony#330, the undraft order, and ceremony#336). Re-vendored all six
manifest files and moved the pin recorded in .ceremony/README.md.
#39 moved the two workflow callers to 0.6.3 but left the doctrine mirror at
0.6.1, so BUILDER.md was missing the three clauses that govern signal-then-ready
(ceremony#330, the undraft order, and ceremony#336). Re-vendored all six
manifest files and moved the pin recorded in .ceremony/README.md.
#39 moved the two workflow callers to 0.6.3 but left the doctrine mirror at
0.6.1, so BUILDER.md was missing the three clauses that govern signal-then-ready
(ceremony#330, the undraft order, and ceremony#336). Re-vendored all six
manifest files and moved the pin recorded in .ceremony/README.md.
#39 moved the two workflow callers to 0.6.3 but left the doctrine mirror at
0.6.1, so BUILDER.md was missing the three clauses that govern signal-then-ready
(ceremony#330, the undraft order, and ceremony#336). Re-vendored all six
manifest files and moved the pin recorded in .ceremony/README.md.
#39 moved the two workflow callers to 0.6.3 but left the doctrine mirror at
0.6.1, so BUILDER.md was missing the three clauses that govern signal-then-ready
(ceremony#330, the undraft order, and ceremony#336). Re-vendored all six
manifest files and moved the pin recorded in .ceremony/README.md.
Closes#39. 0.6.1 has no draft guard in its Forgejo mergeability mapping, so
the sweep marks every draft PR as blocker:conflict (see !38). The guard landed
in ceremony d3b7984, first tagged 0.6.2.
Closes#39. 0.6.1 has no draft guard in its Forgejo mergeability mapping, so
the sweep marks every draft PR as blocker:conflict (see !38). The guard landed
in ceremony d3b7984, first tagged 0.6.2.
Tokens minted by stoke auth login previously got read/write on every
non-admin scope. Default to the reduced set the common issue/PR/repo
commands need (read/write issue + repository, read user + organization),
add --full-scopes to restore the old behavior and --scopes <csv> for a
custom list, and print the granted scopes after login.
Replaces #12, which committed every generated render into the repo. Rebuilt on
a clean branch because merging the original would have written those blobs into
main's history permanently, even with a later commit deleting them.
What changed from #12:
- Keeps docs/DESIGN.md and assets/logo-mark.svg (1.4 kB of vector text).
- Drops ~23 MB of PNG/MP4. They live in the Figma file, which was already the
source of truth and is linked from the doc. stoke's .git is ~23 MB; those
assets would have doubled it, forever.
- Adds the missing "files" whitelist to package.json. There wasn't one, so
npm pack shipped the whole working directory: measured 23.7 MB with the
assets, and it was already shipping the test suite without them. Now 26.7 kB
across 7 files.
The packaging bug is pre-existing and independent of the design work; the
oversized PR is just what made it visible.
61/61 tests pass; `stoke --version` → 1.3.0.
