autonomic-bot 83c183d985 feat(canon): §2.G strip UPGRADE_BASE_VERSION entirely (plausible verified dynamic-base green) ...

Gate satisfied — live: with the pin removed, plausible's upgrade tier resolves base 3.0.1+v2.0.0 via
the same-version step-back (canonical 3.1.0 == head 3.1.0 → newest-older = 3.0.1, NOT the broken
3.0.0) and passes install+upgrade green (level 5/5). The pin is redundant, so removed everywhere:
- meta.py KEYS entry (RecipeMeta field auto-drops; 15→14 keys).
- run_recipe_ci.resolve_upgrade_base override branch + docstrings.
- tests/unit/test_meta.py (count 15→14, dropped None-assert), test_upgrade_base.py (override test).
- docs/recipe-customization.md (regenerated table + mentions), docs/testing.md.
- tests/plausible/recipe_meta.py (pin removed), tests/bluesky-pds (re-enable note → dynamic base).
294 unit tests pass; lint clean.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-06-17 12:31:53 +00:00

..

__init__.py

M4: harness + green install stage (custom-html + Playwright); guaranteed teardown; M4 CLAIMED

2026-05-27 00:23:55 +01:00

abra.py

feat(samever): step back to older base when last-green canonical == head version

2026-06-17 04:24:14 +00:00

browser.py

style: repo-wide lint pass — make the lint gate green again

2026-06-09 21:56:15 +00:00

canonical.py

feat(harness): P1 — single registry-backed meta loader (rcust)

2026-06-10 16:46:58 +00:00

card.py

feat(lvl5): P1 — 5-rung ladder (L5=abra recipe lint) + de-capped level semantics

2026-06-11 07:42:30 +00:00

deps.py

fix(drone-dep): revert _count_deploy=False — dep deploys must count for DG4.1

2026-06-11 21:59:51 +00:00

discovery.py

feat(cfold): canonicalize custom test layout

2026-06-12 16:08:18 +00:00

generic.py

fix(prevb): prune orphan services on upgrade redeploy (head's dropped services); re-add EXPECTED_NA-other-rung test; consume Adversary inbox

2026-06-17 00:29:00 +00:00

http.py

style: repo-wide lint pass — make the lint gate green again

2026-06-09 21:56:15 +00:00

level.py

feat(lvl5): P1 — 5-rung ladder (L5=abra recipe lint) + de-capped level semantics

2026-06-11 07:42:30 +00:00

lifecycle.py

fix(canon): release cold-run app/dep locks before promote (cold-dep self-deadlock)

2026-06-17 10:04:14 +00:00

lifetime.py

feat(harness): P1 lock-lifetime hardening — PDEATHSIG + SIGTERM/SIGALRM teardown funnel + 60-min hard deadline

2026-06-10 04:04:28 +00:00

lint.py

fix(lvl5): lint executor PR-path — abra lint selects+checks out the repo DEFAULT BRANCH; scratch clone of a detached per-run tree has none (FATA, live 400-402), and a stale default would be silently linted instead of the PR head. Force local main AT the tested ref + repoint origin to the scratch itself (offline tag fetch, no drift). Regression test with detached two-commit source proves exact-ref content is linted. 247 unit tests green; real-abra detached-source smoke pass.

2026-06-11 10:56:56 +00:00

manifest.py

fix(gtea): ruff format + check all gtea files and bridge.py

2026-06-15 21:52:01 +00:00

meta.py

feat(canon): §2.G strip UPGRADE_BASE_VERSION entirely (plausible verified dynamic-base green)

2026-06-17 12:31:53 +00:00

naming.py

style(1b): auto-format + lint-clean the whole codebase (RL1)

2026-05-27 20:52:05 +01:00

results.py

feat(lvl5): P1 — 5-rung ladder (L5=abra recipe lint) + de-capped level semantics

2026-06-11 07:42:30 +00:00

screenshot.py

fix(shot): A1 — blank-retry keeps the LARGER frame (retry snapped to temp path, os.replace only if >= first; worse late frame discarded + temp cleaned); regression test [9999,4801]->9999; 207 unit tests pass, lint PASS

2026-06-11 06:24:01 +00:00

sso.py

fix(lint): F821 undefined 'e' in test_scm_configured; shfmt/ruff auto-fixes

2026-06-11 22:17:19 +00:00

warm.py

feat(2w): W1 canonical registry module (WC2) + alerts archived

2026-05-29 02:15:11 +01:00

warmsnap.py

style: repo-wide lint pass — make the lint gate green again

2026-06-09 21:56:15 +00:00