DEPLOY-WINDOWS-IIS was a second copy of the manual IIS procedure that had
diverged from the first: a different MySQL version (8.0, which reached end of
life in April), a different port, a different plugin list, and a profile file
that does not exist. Two runbooks for one procedure means a reader follows
whichever they found, and one of them was wrong. INSTALL-WINDOWS-IIS covers
everything it did plus a preflight step and the subpath method, so the one
section it uniquely had - redeploying a hand-built server - is folded in there,
with the plugin-chain step it was missing and a note to back up first, and the
duplicate is gone. Everything that pointed at it now points at the survivor.
Three ADR statuses said something untrue.
ADR-013 said PROPOSED while half of it had shipped and ADR-014 had been accepted
on top of it. A decision that has been implemented and depended upon is not
proposed, and leaving one that way devalues every other status in the index. The
catalog half is still unbuilt, which is the ordinary state of an accepted
decision: accepted means settled, not delivered.
ADR-016 said ACCEPTED for a design where nothing is built - the endpoint and
permissions it describes do not exist, so a reader goes looking for them. The
status stands, because the decision does; the header now says so plainly and
points at where today's credentials actually live.
ADR-003 and ADR-004 were ACCEPTED with their own Decision lines still opening
"**PROPOSED:**", which reads as though the decision was never taken.
And the dashboard proposal carried Status: ACCEPTED, which belongs to a decision
record. A proposal is a proposal; the contract it produced is the ADR.
The publishability gate caught internal tooling names and developer paths but
nothing site-specific, so roughly sixty leaks reached the wiki: the site name in
ten documents, real fleet hostnames in the collector and GE-Enforce examples, an
internal database name through the whole import guide, imaging-share paths, and
a maintainer's username as the Deciders line of every ADR and inside a generated
curl example.
None of it is a security matter on an air-gapped fleet. It matters because these
pages are read by engineers at other plants, and a document that names one site
throughout reads as that site's notes rather than a product's documentation -
which is exactly what it then gets treated as.
Examples now use neutral hostnames, the site is "the reference site" where the
distinction carries meaning, and ADRs are decided by "ShopDB maintainers". The
gate carries all of these patterns, so the next one fails a build.
Two documents leave docs/ because they were never written for an outside reader.
PROJECT-REVIEW.md is an internal health memo pinned to a commit from July, whose
headline finding (an untracked playbook) has since been fixed - it is history,
and git holds it. PILOT-DEPLOY.md is one site's own cutover runbook, complete
with a "re-measure before publishing" placeholder; it moves next to the loader
it belongs to, in scripts/site_imports/wjf/.
ADR-015 is AMENDED rather than rewritten. Its enforcement section still said
report-only and its backlog still listed hardcodes that are now cleared, which
left the record contradicting itself. The amendment says what changed and why
the report-only period ended; the original text stays, because what the decision
looked like when it was taken is the part worth keeping.
Also corrects llms.txt's response envelope, which had errors at the top level
and pagination at meta.total. Both are nested one deeper, so anything written
against that description read undefined on every error it tried to handle.
An ACCEPTED ADR ended with a stray tool-call fragment - it said nothing to a
reader and everything about how the document was produced, which is not what a
decision record is for. And the cutover playbook carried a real internal IP in a
file otherwise correctly scrubbed to documentation ranges. Both were published.
The publishability gate now catches all three shapes, so the next one fails a
build instead of reaching the wiki.
The adoption guide also told a site to copy two files that ship from nowhere:
the asset reporter and EventSaver are both on the reference site's imaging
share, not in this repository, and the guide never said so - it could not be
followed as written. It now states, per artifact, where each comes from, which
of them are pending a move into this repository, and that EventSaver can be
rebuilt from its one source file rather than taken on trust as a binary.
The scanner has been reporting the same count for weeks, which is what a rule
that only prints becomes. It now FAILS the build, and it looks where the leaks
actually were: PowerShell, the installer, the seeds, generated JSON, the
frontend - case-insensitively, across plugins, shopdb, scripts, deploy, tools.
A line that is deliberate declares itself with an ADR-015-OK marker and a
reason, so the claim is visible in review instead of tolerated in silence.
What it found, fixed here:
- The shadow client wrote one site's ShopDB URL into HKLM whenever the registry
disagreed. At the site it was written for that reads as healing drift;
anywhere else it overwrites the site's own address on every enforce cycle,
and the site cannot win because the cycle repeats. The bay's value now wins,
an explicit -BaseUrl seeds it, and with neither there is nothing honest to
write, so it says so and skips.
- The kiosk dispatcher fell back to one plant's host when HKLM was unset, so a
kiosk elsewhere quietly opened a server it has no business reaching. The
fallback is now this site's site_base_url, baked in at seed time, and the
dispatcher refuses rather than guessing when neither is set. Its legacy
shortcut matcher derives the host from that URL instead of naming one.
- The OpenAPI generator hardcoded a production hostname into every spec it
generated, which then published to a public wiki. The relative mount is the
only server it can honestly name; a site passes its own by environment.
- Placeholders and examples in the UI and the client help offered real internal
subnets and a real production URL. They now use documentation ranges.
Both publication gates - the export scrub and the docs publishability test -
carry the site patterns, which neither did. One plant's hostname, FQDN and
internal networks are out of the documentation and the generated specs.
Comments naming the reference site are reworded rather than deleted: the
reasoning is worth keeping, the plant name is not what makes it true.
Product version to 0.9.0, frontend in lock-step, Unreleased notes moved into a
dated section per ADR-007. The plugin contract stays at 0.16.0: it moved after
v0.8.1 and is already recorded in this release's notes, and nothing since
touched the contract surface.
A minor rather than a patch: collector behaviour changed in ways an integrator
must know about. A reported machine number no longer becomes the PC's asset
number, it builds a controls link instead; a second PC claiming a machine is
treated as a claim rather than a handover; and a backup revision chain is now
per source PC rather than per asset.
Controller and share credentials are currently edited as cleartext into a
script on the SFLD share, one value for the whole fleet, re-applied by
GE-Enforce every cycle. Rotating means editing that file and hoping - nothing
reports which bays picked the new value up - and sites need per-bay variation,
which a manifest scope cannot express because scopes are per PC TYPE. A scope
per bay is a spreadsheet, not a model.
ShopDB therefore owns the definitions and the targeting, and a PC receives a
decided answer rather than rules to evaluate. Targeting uses the axes
GE-Enforce already filters on plus those the asset model knows: hostname
(including -like globs), machine, model, controller OS, PC type with alias
expansion, GE-Enforce profile, and a selector over the stored DNC projection -
"every machine whose DataHost is that share" is answerable from data already
held, and it also answers which machines need a credential at all.
Overlapping matches are guaranteed rather than exceptional, so precedence is
explicit: an integer priority, then scope specificity, then rule id for
stability. The resolver must be able to explain which rule won and why, or
nobody will trust it. It reuses plugins/geenforce/filters.py; there are already
two implementations of this matching logic kept honest by a parity harness, and
a third would drift silently.
Targeting stays separate from detection, mirroring PCTypes versus
DetectionMethod: who gets a credential is a server question, whether it is
already applied is a local one, and conflating them rewrites the secret every
cycle instead of healing drift.
Stored Fernet-encrypted with the key OUTSIDE the database, write-only through
the API and masked on read, fetched with a dedicated credentials.fetch scope so
a leaked collector key does not yield controller passwords, and every fetch
audited.
PROPOSED, not accepted: it makes ShopDB a credential store, and the key becomes
a single point of loss - restore the database without it and every credential
is unrecoverable, which docs/BACKUP-RESTORE.md must state in the same change
that implements this. Registry-only provisioning is recommended as a first step
regardless, since it removes the cleartext from the share immediately and the
client helper is identical either way.
ShopDB is a product several sites adopt, developed at one site against that
site's live fleet, so every feature arrives carrying West Jefferson's
hostnames, networks and share paths. A scan of plugins/ and shopdb/ returns 19
site literals, of which about 8 are a shipped default or a hardcode rather than
a comment: a kiosk URL baked into a PowerShell payload a sister site cannot
override, a WJ UNC path as the backup share default, and device.geaerospace.net
shipped as a SETTING DEFAULT - inherited silently by anyone who adopts.
The mechanisms already existed; the rule did not, so each value landed wherever
was convenient. The rule: a setting with a NEUTRAL default first, a
site-namespaced directory second, seed data third. Blank must be a working
state. Fleet-wide vocabulary such as the gea-shopfloor-* imaging types is not
site data and is out of scope - it already has a per-site override.
Enforced by an addition to check-naming-and-style.sh, which CI already runs, in
REPORT-ONLY mode. Making a 19-item backlog a hard gate the day it lands blocks
every unrelated commit until someone clears it; SITE_LITERALS_ENFORCE=1 turns
it into a gate once the listed hardcodes are done. The ADR carries that backlog,
including that the display FQDN domain is defined three times in three files.
docs/ is excluded from the code bundle and its scrub gate, because it goes to
the GitHub wiki instead - via a generator that has no gate at all. So the one
part of the repository written in prose, by people, about internal
infrastructure, was the one part nothing checked.
What was reaching a public wiki: the internal git server's URL and hostname,
.gitea workflow paths, developer home directories in the GE-Enforce cutover
reference, and a dev database root password inside a copy-pasteable command in
the import guide.
All replaced with neutral equivalents. tests/test_docs_publishable.py is now the
gate, at the source, in CI - a wiki page cannot be un-published, so catching this
after the fact is not good enough.
PROJECT-REVIEW.md also referred to internal tooling by name throughout; those
references are generalised. It remains an internal candid assessment of this
project that is nonetheless published, which is worth a separate decision.
A lean site still gets every plugin's tables from the shared core Alembic
baseline. prune-schema drops the tables of plugins not installed on this
site, leaving core + chosen-plugin tables, with no edit to any released
migration (the relocate-into-plugin-baselines alternative would mean
rewriting ~15 released core migrations for a cosmetic gain - see ADR-014).
- shopdb/plugins/cli.py: prune-schema command. Dry-run by default; --yes to
execute; refuses non-empty tables without --force. Drops by table name (no
plugin import) so it works on a lean image. MySQL: private AUTOCOMMIT engine
(db.engine's pooled connections sit idle-in-transaction in a CLI context and
would deadlock the DROP on a metadata lock). SQLite: db.engine, restoring the
prior foreign_keys pragma so the StaticPool connection is not left changed.
- tests/test_plugin_prune_schema.py: drop-only-not-installed, full no-op,
refuse-non-empty, force-drops-non-empty.
- docs/DEPLOY.md: lean provisioning step after upgrade-all.
- ADR-014 ACCEPTED; index updated.
Verified on MySQL: full install then prune = no-op (86 tables); lean install
(machines+printers) then prune drops the other 19 plugin tables; second run
no-op. Full suite 1077 passed.
Cross-plugin FK blocker ADR-013 cited is already resolved: the FKs into
machines were held only by dead legacy tables (machinerelationships,
printerdata, installedapps, communications.machineid) that existing
migrations 7a01/7c01 already drop. No live plugin table hard-FKs another
plugin. Schema-lean is unblocked.
Enabling change: create_plugin_tables now skips already-existing tables
(idempotent) so a plugin anchor can create its tables on a fresh lean
install and no-op on a database that has them from the pre-cutover
baseline. The load-bearing baseline lift is staged as ADR-014 Phase 2.
Design record for distributing optional plugins across GE sites: a small
mandatory core plus a catalog of optional plugins, packaged as signed
versioned artifacts, served from a transport-agnostic read-only shelf (a
SharePoint-synced or sneakernet folder - untrusted either way because every
decision-bearing byte is signed), verified at adopt AND at every load and
migrate. Lean per-site builds stage only chosen plugins into the backend
image and SPA bundle.
Status PROPOSED. Grounds the design in the real loader/contract/migration/
frontend code and records defects to fix along the way (upgrade-all
migrating unadopted folders, enable-without-install, reverse-dep checks
blind to unloaded plugins, missing cycle detection and dependency closure,
hardcoded plugin imports in the SPA router). Honest on scope: the frontend
re-org is the long pole (one core-router change plus per-plugin relocation),
not a mechanical move. Phased 0-5 with schema-lean and runtime-JS delivery
explicitly deferred.
From the Fable/Opus documentation audit (8 confirmed + verified
lab-drift the run's session limit had cut short):
- HIGH: the lab's kiosk _kiosk_find_item block showed the pre-stage-17
row-id resolver as current; replace with the shipped gagelabtag /
numeric-tail resolver, fix the stale 'resolved by row id' prose and
the 'stage-7 code is corrected' note.
- MED: the badge _external_lookup block used dict-only row access that
breaks on a tuple cursor; use the tuple-or-dict form shipped. Split
'&&' command chains (fail in PowerShell 5.1) in the lab.
- LOW/link: the Windows note's [DEVELOPMENT-SETUP] link dropped the .md
and 404'd in four docs; fix. Correct the stage-6a->16a comment and
the lab-stage tag range (..16 -> ..17).
- Leaks: drop /home/camp path from ADR-006, the internal gitea host
from PLUGINS.md.
- Windows: add an mklink junction note for the external-plugin symlink
dev loop.
- CI: prime root to mysql_native_password so pymysql connects to the
MySQL 8 service without the cryptography package (and its kit wheel).
Import-surface docs and docstrings describe the automation as a
migration script; status-doc references in CHANGELOG/ADR-009/ROADMAP
point at repo files. Screenshot/verify tools write to /tmp/shopdb-shots
(created on import) instead of a machine-specific directory.
Four data-only hooks on BasePlugin (get_settings_cards,
get_asset_panels, get_map_overlays, get_asset_presentation) with a
GET-only /api/pluginui consumer surface copying the dashboard-widgets
semantics. Pilots: warranty declares its asset panel; measuringtools
supplies its settings card, presentation, and calibration overlay -
the last hardcoded settings-nav entry is now hook-sourced. Generic
renderers for panels/overlays/presentation deferred per the ADR's
incremental adoption plan (documented in CONTRACT-STABILITY.md).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Records the approved domain rename (equipment plugin/tables/API ->
machines) and the lossless retyping of the models catalog (legacy
machinetypes -> modeltypes), with migration placement per ADR-008.
Implementation lands in the following commits.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
ADR-010 (PROPOSED): four data-only frontend hooks following the
get_reports precedent - get_settings_cards, get_asset_panels,
get_map_overlays, get_asset_presentation - batched as a future contract
0.7.0. Warranty proves asset panels first, measuringtools the rest;
accept only after both run on the hooks.
Display-label swap Equipment -> Machines across nav, list, form,
detail, search, map editor, and settings copy (incl the API-failure
fallback nav). Identifiers unchanged: plugin name, tables,
/api/equipment, asset type, and routes all stay.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Framework:
- Per-plugin Alembic migration chains (ADR-008): every bundled plugin
carries its own chain with a stamp-only anchor at the ownership cutover;
new plugin schema lands in plugins/<name>/migrations/, never the core
chain. Deploys add flask plugin upgrade-all. Fixed a latent bug in the
shared alembic template (engine URL resolution) and taught the metadata
filter to include FK-referenced core tables.
- Frontend plugin route gating (ADR-009): plugin routes carry meta.plugin;
a disabled plugin's pages redirect to the dashboard via a cached,
fail-open check against the new public GET /api/plugins/enabled.
- get_reports() plugin hook (contract 0.5.0 -> 0.6.0): plugins contribute
report cards; warranty and toner cards moved off the hardcoded list.
Reports:
- Hub grouped by category with search; inline reports render at the top,
are URL-backed (?report=id, back-button and deep links work), expose
their server-side filter params as controls, and export CSV. Warranty
and Toner pages gained CSV export.
- Deleted the dead legacy Warranty Status report (always-zero buckets
from a retired column).
Theming and fonts:
- Inter (variable) bundled locally via @fontsource, replacing the Google
Fonts Roboto import - air-gapped installs now render correctly; tables
use tabular numerals.
- Optional brand_primary_dark_color, brand_accent_color,
brand_sidebar_color settings applied to CSS vars at bootstrap.
USB frontend repair (views were reading a dead legacy shape):
- List/detail/form and the employee profile USB panels remapped to the
real API shape (device_id/device_desc/checkinoutlog); employee panels
now use /usb/checkouts endpoints; external-mode /usb/checkouts/active
honors the badge filter; dead client methods pruned.
Also: warranties list page no longer requires login (matches app
convention); collector doc rewritten with a GE-Enforce integration guide
and paste-ready PowerShell reporter; ADR index and CHANGELOG updated.
Verified: 323 tests pass, naming/style green, frontend builds, plugin
migration dry-run green on scratch MySQL.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Make the app distributable to other GE Aerospace sites (one self-hosted
instance per site, ADR-004). GE values remain the shipped defaults; every
site-specific behavior is now a Setting an admin can change in the UI.
Settings-driven site config:
- Branding: site/QR/badge logos, favicon, primary color (upload endpoints
mirror the map-blueprint pattern; new Settings > Branding section).
- ServiceNow: search/incident/change URL templates ({ticket}), ticket
prefixes, enable toggle. Defaults point at the current
geaerospaceqa.service-now.com global search. Disabled = plain-text tickets.
- Employee-id regex (employeeid_pattern), printer hostname template,
QR label targets (qr_target_printer / qr_target_usb, blank = asset page,
else URL template with placeholders), usb_label_style (barcode|qr).
- West Jefferson floor-plan PNGs removed from the tree; generic placeholder
ships as the map default and sites upload their own blueprint.
Security closeout:
- dashboarddefaults writes now require admin.
- Collector: generic error messages (no str(exc) leak); API key accepted
via X-API-Key header only (BREAKING: querystring api_key removed).
- IP-based login rate limiting (AUTH_RATELIMIT_* knobs) atop account lockout.
- Setting.set() creation race fixed (IntegrityError retry).
Release engineering and docs:
- __version__ 0.5.0 (distinct from __contract_version__, ADR-007),
CHANGELOG.md, Gitea Actions CI config, frontend version aligned.
- One wizard-first install story across README/DEPLOY; new CONFIG.md,
UPGRADE.md, BACKUP-RESTORE.md; CLAUDE.md and ROADMAP de-staled.
- Dockerfile multi-stage build now bundles the frontend; compose binds
MySQL to 127.0.0.1; stale database/schema.sql and one-off SQL removed.
Debt and fixes:
- .query.get() -> db.session.get() sweep; datetime.utcnow() removed
(naive-UTC via timezone-aware now); users.py on authz decorators.
- Fixed 4 stale tests (slides feed shape, shopfloor splitperemployee,
plugin contract purity) and the USB label page field mapping (both usb
modes emit the cmmc shape: device_id/device_desc).
- Health endpoint reports the real version.
248 tests pass; naming/style check green; frontend builds; fresh-DB
flask db upgrade + seeds verified; QR targets verified by decoding
rendered codes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Feature work from the 2026-07 session:
Settings IA
- Replace the flat 27-card settings hub with a persistent two-pane shell
(SettingsLayout.vue): grouped, searchable left rail + content pane.
- Nest all settings/* routes under the shell via router post-processing;
shared nav catalog in settingsNav.js. Group by asset class (PCs, Printers,
Equipment, Network) so per-type settings stop scattering.
Custom fields (core)
- customfields + customfieldvalues tables (migration 7d14), CRUD API at
/api/customfields, per-asset value get/save.
- Settings management page + reusable CustomFieldsSection (detail) and
CustomFieldsInputs (form) wired into all four asset types.
Warranty (new plugin)
- plugins/warranty: warranties + warrantyassets (migration 7d15), derived
coverage status, provider abstraction (manual now; Dell/Lenovo/HP stubs).
- API CRUD + per-asset panel + report buckets; WarrantyPanel on all four
detail pages; Warranties management page; Warranty report + Reports card.
- Seed warranty.* permissions.
Printer drivers
- printerdrivers table (migration 7d13) linked to printer models; drivers now
surface on the matching printer's detail page.
Other
- PCDetail rebalanced (Network + Status + Warranty + custom fields on the right).
- Rename PCs list "Features" column to "Remote Access"; fix badge hover underline.
- Drop equipment islocationonly field.
- Centralize asset-type label/route maps into utils/assetTypes.js.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- USBLabelBatch.vue: recolor controls/selection/cell borders to the app CSS
variables (was hardcoded #667eea/#28a745/#dc3545) and add
`print-color-adjust: exact` so barcodes print even with "Background graphics"
off. (Barcodes render as vector SVG, so no canvas->img change needed.)
- ADR-004: document the resolved migration strategy - single core Alembic chain
(`flask db upgrade`), bundled plugin schema folded into core (7c04), resolving
the Phase 7B per-plugin-chain conflict footgun. External plugins may still
ship their own chains.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Architectural pass from the skill review ("plugin is the product" boundary).
Core no longer imports plugin models at module load (was a hard import-time
dependency that broke core if the computers plugin was absent/disabled):
- collector.py, applications.py, reports.py: lazy + guarded imports of the
computers plugin models. Endpoints that need install-tracking now return 503
when the plugin is absent instead of failing at import.
Search honors runtime enable/disable:
- search.py: _require_enabled(name) raises ImportError for a disabled plugin,
so each plugin-scoped block skips it (a disabled plugin's rows leave search).
- Replace hardcoded root/rootpassword employee-DB connection in _search_employees
with the shared env-backed employee_connection helper.
Plugin hooks (integrating-plugin-hooks: every hook needs a consumer):
- get_dashboard_widgets: add the consumer GET /api/dashboard/widgets (5 plugins
already implemented the hook; it had none). Skips disabled, isolates in prod.
- get_searchable_fields: REMOVED. Zero plugins implemented it and there was no
consumer; global search is a core concern over the asset model. Contract
reduction, __contract_version__ 0.3.0 -> 0.4.0.
Docs/contract: PLUGIN-HOOKS.md (widgets consumer note, searchable-fields removal,
0.4.0), PLUGIN-QUICKSTART.md, ADR-001 hook list. Tests: widgets endpoint
aggregate + disabled-skip; contract tests for the removed/added hooks.
151 tests pass, naming/style green, app boots all 6 plugins.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Addresses findings from a 6-lens review against the project skills
(defining-asset-contract, enforcing-plugin-contract, hardening-flask-config,
integrating-plugin-hooks, pinning-flask-behavior, simplifying-python).
Security (hardening-flask-config):
- Load per-plugin COLLECTOR_API_KEY_<PLUGIN> from env in create_app. from_object
only copies class attributes, so per-plugin keys (ADR-006) were dead in real
deploys and silently fell back to the shared key.
- EMPLOYEE_DB_USER/PASSWORD no longer default to root/rootpassword (no safe
default for a secret; unset fails loud). Documented in .env.example + DEPLOY.md.
- COLLECTOR_API_KEY + per-plugin + EMPLOYEE_DB_* added to .env.example/DEPLOY.md.
Hook isolation (integrating-plugin-hooks):
- collector _collector_plugins and dashboard get_navigation now re-raise in
dev/test and log+isolate in prod, instead of silently swallowing a broken
plugin hook.
Plugin loader (enforcing-plugin-contract):
- enable_plugin/install_plugin read dependencies+version from the manifest
instead of instantiating the plugin class.
- _register_plugin_components rejects a second plugin claiming an already-used
api_prefix (reset per app in init_app).
Tests (pinning-flask-behavior):
- test_identifiers.py: gauge/maintenance round-trip on computer/printer/network
create+update; per-type seed yields the 12 identifier keys.
- contract tests for apply_collector_payload presence + schema-declarers-implement.
- security tests for per-plugin key env loading + no employee-db password default.
Docs/contract sync (defining-asset-contract):
- PLUGIN-HOOKS.md documents apply_collector_payload; stale 0.2.0 -> 0.3.0.
- ADR-006 documents apply_collector_payload + single-dispatch rationale.
- ADR-001 enumerates the expanded shopdb.api import surface.
Simplify (simplifying-python):
- De-duplicate the 21-entry settings defaults: shared build_default_settings()
used by both the /settings/seed route and the CLI (were drifting copies).
- Remove dead AssetStatus import + redundant AssetType local import in computers
plugin; comment the statusid=1 collector default.
153 tests pass (was 145), naming/style green.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Migration runner ready and a sister site can deploy from a clean
checkout with one .env file.
ADRs relocated (migrations/adr/ -> docs/adr/):
- migrations/ is now Alembic territory, not docs.
- All cross-references updated: CLAUDE.md, docs/PLUGIN-HOOKS.md,
docs/PLUGIN-QUICKSTART.md.
Alembic initialized (migrations/):
- env.py, script.py.mako, alembic.ini copied from Flask-Migrate
templates so `flask db migrate` and `flask db upgrade` work without
a one-time `flask db init` (which would clash with the existing
migrations/ directory).
- Baseline migration generated via autogenerate, captures all 47
tables (core models + 6 plugins) as the upgrade target. Ready for
per-site `flask db upgrade` from an empty schema.
Deploy artifacts:
- Dockerfile: python:3.12-slim base, gunicorn server, non-root user,
healthcheck against /api/auth/login. Single image bundles all six
plugins; sites enable via `flask plugin install <name>`.
- docker-compose.yml: MySQL 8 + API container, healthcheck-gated
startup, env-driven secrets that fail loud on missing values
(`${SECRET_KEY:?}` form).
- .env.example: full env-var inventory with comments. Calls out
required vs optional. Matches what ProductionConfig.validate
enforces.
docs/DEPLOY.md:
- Step-by-step per-site runbook: clone, configure .env, bring up
stack, run migrations, seed reference data, install plugins,
create admin, front with TLS, backups, updates.
- Common-issues table.
- Cross-links to ADR-004 (per-site rationale), ADR-003 (plugin
distribution), and the config source.
Skills:
- migrating-asset-schema: Alembic + one-shot data migration policy.
Rules: additive first, renames are three steps, destructive ops
need rollback, equipment migration filter per ADR-001 + ADR-005.
- hardening-flask-config: production validation, CORS allowlist
policy, JWT cookie hardening, per-site deploy isolation per ADR-004.
CLAUDE.md updated to reflect the post-Phase-5 state. No tests added
this commit; the Alembic baseline is exercised by the existing
db.create_all-based test suite (tests do not touch the migration
runner; that's by design until per-plugin migrations land).
Test count unchanged: 101 passing.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
