Age stale pre-delivery attempts to failed and in-delivery attempts to uncertain without provider I/O. Notify each owner with content-free guidance and keep recovery idempotent across replicas and restarts.
Require tenant-owned jobs, explicit confirmation, canonical request IDs, and rate limiting before provider delivery. Persist sent correspondence with a content-free idempotency ledger, and never retry uncertain outcomes automatically.
Gmail and Graph request explicit send consent and classify provider rejection separately from uncertain transport failure. IMAP remains read-only; no send API is exposed.
CV upload now returns 202 with an owner-scoped operation instead of holding the request through parsing. Existing review approval remains required.
BREAKING CHANGE: profile-cv upload responses use the durable operation contract.
Production's SQLite-shaped Jobs table lacks AUTO_INCREMENT, causing the legacy opportunity backfill to abort startup. Reuse the existing idempotent MariaDB primary-key repair before backfill runs.
Route public health checks to the API, backfill and synchronize job opportunities, stabilize SPA smoke tests, and document operator-only production steps.
- consolidate API ownership and remove dead vendor code
- add Stripe billing, learning paths, and public CV hardening
- add migration, recovery, security, audit, and browser gates
Production GET /api/cv/outline returned 500 "Unknown column
'c.LongTailJson'". CareerProfiles is reconciler-owned, but the
reconciler's CREATE TABLE (both the SQLite and MySQL branches) only
listed Id, OwnerUserId, ProfileJson, Version, CreatedAtUtc, UpdatedAtUtc.
LongTailJson was added to the CareerProfile model in Phase 3 but neither
CREATE was updated and no column-repair existed, so:
- existing databases (prod): the MySQL CREATE is guarded on
!HasMySqlTable, so it never runs once the table exists, and nothing
adds the column -> LoadStructuredAsync selects a column that isn't
there.
- fresh databases: the CREATE itself omitted the column, so even a brand
new MariaDB/SQLite was missing it. The 420 tests never caught this
because they build tables from the EF model, not the reconciler DDL.
The release audit missed it because it never exercised /api/cv/outline.
Add LongTailJson to both CREATE statements and add an additive repair
(EnsureColumn / EnsureMySqlColumn) for existing tables. DEFAULT ''
backfills existing rows and matches the non-nullable model property.
This is the sanctioned reconciler repair path, not a manual ALTER, and
preserves existing data (ADD COLUMN is non-destructive).
Verified on a real MariaDB 11 container: an existing 6-column
CareerProfiles gains LongTailJson on startup (repair path), a fresh DB
gets it from the CREATE (longtext), and GET /api/cv/outline returns 200.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The reconciler's IX_JobApplications_OwnerUserId_FollowUpAt was declared as
(OwnerUserId(191), FollowUpAt) with no prefix length on FollowUpAt. But
FollowUpAt is `text` on MariaDB -- JobApplications is migration-owned and
the migration was scaffolded against SQLite, which stores DateTimeOffset
as TEXT. A text column cannot be indexed without a prefix length, so this
index failed the 3072-byte key check on EVERY MariaDB boot, was caught by
TryCreateIndex, and was silently skipped -- leaving the follow-up reminder
query (OwnerUserId + FollowUpAt) unindexed.
Two consequences, both real:
- the index the code intends to create never existed on MariaDB
- every healthy boot logged "Specified key was too long", which
deploy/first-production-deployment.md lists as a STOP-AND-ROLL-BACK
signal -- so an operator following the runbook could abort a good deploy
The author already handled the identical problem for the longtext Status
column one line below with Status(50). Apply the same fix: FollowUpAt(20).
ISO-8601 date strings sort lexicographically, so a 20-char prefix
("YYYY-MM-DD HH:MM:SS") keeps the index useful for the reminder scan.
Verified on a fresh empty MariaDB 11 container: the index is now created
(both key parts present), zero "Specified key was too long" lines, zero
skipped indexes, zero unhandled exceptions, 42 tables, app healthy. This
was the only unprefixed text column in any reconciler composite index --
the datetime columns on reconciler-owned tables are datetime(6). SQLite is
unaffected (its CREATE INDEX has no key-length limit).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
/health read the APP_VERSION environment variable directly, but
docker-compose passes App__Version, which binds to the App:Version
configuration key. The variable under that name never existed in the
container, so the endpoint always reported "unknown".
Read App:Version through IConfiguration, the approach AdminSystemController
already used for the same value. The resolution rule (configured version,
else assembly version) moves to a shared BuildMetadata helper rather than
being written twice; AdminSystemController now calls it, so the admin page
and /health cannot drift apart.
Local development is unaffected: nothing sets App:Version there, and the
assembly-version fallback still applies.
Tests pin the configuration KEY, not just the behaviour, including that an
App__Version environment variable binds to App:Version. The original bug
failed silently, so a behavioural test alone would not have caught it.
Verified against a running backend: App__Version=9.9.9-test reports
9.9.9-test; unset reports the assembly version rather than "unknown".
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The timeline could interpret InterviewScheduled, InterviewCompleted,
OfferReceived and FollowUpCompleted, but only StatusChanged and FollowUpSet were
ever written, so those branches never rendered.
Events are now derived from the status TRANSITION in one shared emitter rather
than at each call site, so the two status-change boundaries in
JobApplicationsController cannot drift apart and a third would get the behaviour
for free. Both boundaries now call it instead of hand-writing the StatusChanged
block.
Deriving from the transition rather than the resulting state is what prevents
duplicates: one user action produces at most one lifecycle event, re-saving an
unchanged status produces none, and reaching an offer twice records it once.
Moving an application backwards is treated as a correction, not a completed
interview, so only a forward move out of an interview stage counts. An
Interview to Offer move reports the offer, which is the thing the user cares
about.
Completing a follow-up checklist item emits FollowUpCompleted, guarded on the
same transition rule so re-saving a done item stays silent. The task itself
remains a checklist item — this only records that it happened.
No new history store: every event is a JobEvent row, which stays the single
source of application history.
393 backend tests pass, including timeline rendering of the emitted events.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Interview generation saw only the profile and the advert, so it produced generic
questions. It now also receives what the workspace already computed: seniority,
employment type, key requirements and advert technologies from the job analysis,
plus the match score, the skills the candidate demonstrably has, the most
relevant experience and projects — and above all the gaps, which is exactly what
an interviewer probes.
No second pipeline. The context comes from ApplicationIntelligenceService, which
is deterministic and read-only, so this adds no AI call and cannot alter user
data. Generation still runs through AiWorkspaceService and is still appended to
AiInteraction.
The dependency is optional, so existing constructions keep working and a missing
intelligence service degrades to the previous prompt instead of failing. Only the
interview module is affected; job-analysis, career-match, cover-letter and
application-review assemble exactly as before.
Suggestion-only is unchanged and now pinned by tests: generation adds an
AiInteraction and nothing else, creates no InterviewPrepItem, leaves existing
prep items and the CareerProfile untouched, and refuses another user's
application. Context is scoped to the requesting user, so another user's profile
is never scored in.
379 backend tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 5.5. Completes the lifecycle after submission: prepare, communicate, chase.
Interview preparation gets a durable, user-owned store. There were already two
per-application AI stores, InterviewPrepNote and AiWorkspaceNote, but both are
caches that regenerate when their context signature changes — anything a user
typed into them would eventually be overwritten. InterviewPrepItem is the side
nothing regenerates, covering company research, technical notes, behavioural
answers, STAR examples and the user's own questions in one table, because those
categories differ only by label and adding one must not need a migration. Each
item records whether the user wrote it or accepted a suggestion, and an
IsPrepared flag makes the section double as the preparation checklist.
Generation stays in the existing AiWorkspaceService "interview" module, appended
to AiInteraction as before. A suggestion is history until the user adds it as a
prep item; opening the section generates nothing.
Follow-up reuses what exists rather than adding a tracker. The date is
JobApplication.FollowUpAt, the same field RulesEngine and the reminder hosted
service already act on, so reminders keep working with no new wiring. The task
stays an ApplicationChecklistItem in the follow-up category — the section counts
open tasks without owning them. The record is a FollowUpSet JobEvent, the same
type the rest of the app emits.
Communication is untouched: Correspondence already owns recruiter contacts,
history and notes, and the workspace already mounted it.
The timeline interpreter learned five more types — InterviewScheduled,
InterviewCompleted and OfferReceived as milestones, FollowUpCreated and
FollowUpCompleted as routine, deliberately outside the milestone spine so it
stays a summary of what actually happened. JobEvent remains the history source.
InterviewPrepItems is reconciler-owned with a no-op migration, guarded on
JobApplications, and verified on a fresh MariaDB 11: int AUTO_INCREMENT primary
key, varchar owner and title, tinyint flag, datetime(6), composite index inside
the key limit.
371 backend tests, 128 frontend tests, Release build and the production build all
pass locally.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 5.4. Connects the career outputs a user already has to one job
application, without building a second copy of any of them.
The flow is strictly one-directional — CareerProfile -> CvVariant ->
application output — and nothing writes back up. No code path in this phase
touches CareerProfile or its children.
CV integration re-points rather than duplicates. GET/PUT /{id}/cv attaches one
variant to an application via CvVariant.JobApplicationId; replacing detaches the
previous variant instead of deleting it. Creating, duplicating, editing, theming,
previewing, exporting PDF and version history all stay in the existing CV
builder, which the section links into. There is no second CV system.
Tailoring composes the Phase 5.3 analysis and match into skills to highlight,
experience to prioritise, projects to emphasise, keywords to include and gaps to
address. Deterministic and advisory: it says what the user could emphasise and
the user edits the variant themselves. Nothing auto-applies.
Cover letters gain the history they were missing. JobApplication.CoverLetterText
stays the current text with its API contract unchanged; CoverLetterVersions
records what it used to be, so an AI rewrite is never destructive. Restore is
additive — the old text comes back as a new version, so what you restored from
still exists. Source and AiAction record whether the user wrote a version or
approved it from a suggestion, and an AI generation only becomes a version once
the user saves it.
Documents are untouched: the existing Attachment system already covers CV, cover
letter, certificates and portfolio files with a Purpose field, so the workspace
mounts that component rather than adding a second upload path.
CoverLetterVersions is the only new table — reconciler-owned, no-op migration,
guarded on JobApplications, and verified on a fresh MariaDB 11: int
AUTO_INCREMENT primary key, varchar(255) owner, datetime(6), composite index
inside the key limit.
360 backend tests, 115 frontend tests, type check, Release build and the
production build all pass locally.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 5.3. Three read-only reads that answer "how suitable is this job", "how
does my experience match", "what am I missing", "what happened previously".
Timeline (GET /{id}/timeline) is an interpretation layer over JobEvent, which
stays the source of historical truth. Each row gains a readable summary, a
category and a milestone flag; events group by day. Milestones are returned
unfiltered, because narrowing the detail must not hide what actually happened.
Job analysis (GET /{id}/analysis) extracts role, company, location, employment
type, seniority, salary, technologies, skills, responsibilities and keywords
from the advert, reusing the existing SkillTagger so the vocabulary matches the
job importer. It also reports what the advert does NOT say, which is usually the
more useful half.
Career matching (GET /{id}/match) feeds the master CareerProfile into the same
JobCvMatchService the CV builder uses, so one application scores identically
whichever surface asks. It returns the score, matched and missing skills, and
which experience and project entries are the evidence for each match.
All three are deterministic and own no data — no new table, no new column, and
nothing writes to the CareerProfile, a CvVariant, or the JobApplication. The AI
narrative stays where it already was, in AiWorkspaceService's job-analysis and
career-match modules, generated only when the user asks and versioned by the
append-only AiInteraction history. Opening a section costs nothing and changes
nothing.
Frontend adds Timeline, Analysis and Match sections to the workspace, sharing
one loader so loading, empty and error states are consistent. The deterministic
answer renders first, with the AI panel below it.
345 backend tests, 104 frontend tests, type check, production build all pass
locally.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A completely empty MariaDB database could not start: the reconciler assumed
migration-owned tables already existed, and migrations assumed reconciler-owned
tables already existed. Neither could go first. Existing databases worked, so
only fresh installs were affected.
Startup is now an explicit sequence: connect, reconcile, migrate, reconcile,
start. The reconciler runs twice because neither position alone works — pass 1
repairs legacy schemas and creates the reconciler-owned tables that migrations
reference, pass 2 picks up everything that could not exist yet on a fresh
database. Every statement is existence-guarded, so the second pass is a no-op
scan on a correct database.
Untangled the overlapping ownership:
- RuleSettings is migration-owned. The reconciler also created it, which made a
clean install fail with "Table 'RuleSettings' already exists". It now only
seeds the default row, and only once the table exists.
- The six CareerProfile child tables are reconciler-owned. Their migration was
scaffolded against SQLite and indexed an unbounded longtext OwnerUserId, which
exceeds MariaDB's 3072-byte key limit; it is now a no-op and the reconciler
carries correct per-provider DDL. OwnerUserId and ItemKey are bounded to
varchar(255) in the model so the index fits.
- Reconciler tables that reference another table are guarded on their parent, so
pass 1 skips them on an empty database instead of failing on the foreign key.
- All index creation goes through one EnsureMySqlIndex helper, guarded on table
existence as well as index existence. This removes ten copies of the raw block
that crashed on a missing table.
- The DbContext-owned connection is no longer disposed by the reconciler, and
Open() is guarded on connection state, so the second pass can reuse it.
Verified against MariaDB 11 and SQLite: empty MariaDB (40 tables, starts),
restart on the populated database (idempotent, rows preserved), empty MariaDB
via the Docker image, fresh SQLite (42 tables), and an existing partially
migrated SQLite dev database (34 tables upgraded to 44 with all 13 applications
and 8 companies intact). 329 backend tests pass in Release.
Ownership rules, startup order, fresh install and production upgrade are
documented in docs/infrastructure/database-ownership.md.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Evolve the existing readiness workflow into one persisted, user-controlled
checklist rather than adding a second tracker.
ApplicationChecklistItem records only completion state and user intent. Each
default system item carries a stable SystemKey and an AutoSignal — the same
signal /readiness already computed — and re-syncs on every read: a satisfied
signal auto-completes the item, a reverted signal reopens it, and a manual tick
always wins. Users can add, reorder, dismiss and delete.
Readiness is refactored into a projection of the checklist (score = completion
percentage, completed/missing = live items by status). Its DTO shape and the
workflowSignal/reminders health view are unchanged, so no API contract breaks.
The workspace's next recommended action now comes from the first pending
checklist item in category priority order (preparation, submission, follow-up,
interview, custom), replacing the parallel ruleset — so the overview can never
recommend something already ticked off, and a user's own task can be next.
The table follows the established MariaDB-safe path: the scaffolded migration is
a no-op and the idempotent reconciler owns the DDL for both providers. Verified
on MariaDB 11 — auto_increment PK, varchar/datetime(6)/tinyint(1) columns, both
indexes inside the key limit, cascade delete, unique system key per application,
and NULL system keys not colliding for custom items.
329 backend tests, 94 frontend tests, type check, production build and both
Docker builds pass locally.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Every JobApplication gets a dedicated workspace at /applications/{id} — a
surface, not a new data store. It owns no data and duplicates none: CV comes
from the Phase 4 CvVariant lens, analysis/match/interview from the existing
AiWorkspacePanel, documents from Attachments, communication from
Correspondence, activity from JobEvent, stage semantics from JobPipeline. No
career data is copied and nothing here writes.
- GET /api/jobapplications/{id}/workspace: one aggregate read (role, company,
stage, dates, attached CV variant, cover letter, documents, AI history,
recent activity) replacing the page fanning out across endpoints
- Next recommended action: ordered rules answering "what do I do next?", the
core product principle for this phase
- ApplicationWorkspacePage: left nav + linkable ?section=, reusing the existing
component for each domain; later-milestone sections say so rather than faking
- Entry point from the job dialog via an optional onOpenWorkspace callback —
the dialog must not depend on router context (it is mounted without a
<Router> in several suites), so the caller owns navigation
- 8 backend tests (aggregate, CV variant surfacing, counts, activity ordering,
next-step rules, tenant scoping)
Local: 314 backend, 88/88 frontend (31 suites), tsc clean, production build ok.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Deploy failed on prod (MariaDB) while the test job was green: backend startup
threw during Database.Migrate(), so deploy.sh's post-deploy health check exited.
Root cause: AddCvVariants/AddAiInteractions were scaffolded against SQLite, so
they bake SQLite type names into their DDL — DateTimeOffset emits `TEXT`, bool/int
emit `INTEGER`, and the PK gets no AUTO_INCREMENT. Run against MariaDB that yields
a structurally wrong table, and the composite index over a TEXT column then trips
"ERROR 1071: Specified key was too long; max key length is 3072 bytes". SQLite
accepts all of it, which is why local/container verification passed.
Fix, following the pattern already used for CareerProfiles/AiWorkspaceNotes:
- both migrations become no-ops; the three tables are reconciler-owned
- reconciler provisions them idempotently per dialect (MySQL: varchar/int
AUTO_INCREMENT/datetime(6); SQLite: CREATE TABLE IF NOT EXISTS)
- DropMalformedMySqlTable rebuilds a half-built table left by the failed
migration, guarded on row count so a table with ANY rows is never dropped
- bound the indexed string columns with HasMaxLength so the model matches
Verified against a real MariaDB 11 container: reproduced error 1071, then
confirmed the corrected DDL yields auto_increment PKs, varchar/datetime columns
and all previously-failing indexes. 306 backend tests green; SQLite container
starts clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 5 backend. A unified AI Workspace for each application, orchestrating the
five suggestion modules through the existing ISummarizerService provider
abstraction and storing every generation as append-only history (AiInteraction)
so outputs can be reused, compared, and deleted — distinct from the existing
AiWorkspaceNote cache (one row, overwritten).
Modules (all suggestion-only, "never invent facts" guardrail, never mutate the
profile/variant/application): job-analysis, career-match, cover-letter (6 modes),
interview, application-review. Each builds a prompt from the job + master profile
text and returns markdown.
- Models/AiInteraction.cs + migration AddAiInteractions (verified on container)
- Services/AiWorkspaceService.cs (prompts, history, delete)
- Controllers/AiWorkspaceController.cs (/api/jobapplications/{id}/ai:
generate, history, delete, modules+provider)
- 7 tests (store, history filter/order, delete, mode normalization, unknown
module, empty output, tenant scoping); 306 backend green
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 4.5 backend enablers.
- Themes (priority 3): AtsFriendly flag on single-column themes (surfaced in
GET /api/cv/themes), print-quality page-break rules (entries never split
across a page; headings stay with content; widow/orphan control), darkened
the creative sidebar for AA contrast.
- Rich text (priority 1): bullets/summary support **bold**, *italic*,
__underline__, [text](url) via a safe inline pass — everything is HTML-escaped
first, so no user tag can survive; only the whitelist emits markup.
- Entry ordering (priority 1): CvSectionSetting.ItemOrder reorders entries
within a section by ItemKey, never touching the master profile.
- Outline API: GET /api/cv/outline returns the master profile as sections+entries
with ItemKeys, so the Content tab can render editable per-item rows.
- 3 new tests (22 total in the builder suite).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 4 foundation. A CvVariant is a lens over the master CareerProfile
(section order/visibility, per-item overrides keyed by ItemKey, theme +
builder settings) — it references career data, never duplicates it. One
renderer (ThemedCvRenderer) draws every theme; a theme is pure data
(CvThemeCatalog, 8 professional themes), so adding a theme needs no renderer
change. Autosave version history + non-destructive restore, public CV via
/api/public-cv/{slug} (anonymous, noindex, filter-bypassing owner load), and
an AI-assist endpoint reusing the existing provider abstraction (suggestions
only, never auto-applied).
- Models: CvVariant/CvVariantVersion, CvVariantSettings, CvTheme + catalog
- Services: CvVariantResolver (profile+lens -> render model), ThemedCvRenderer,
CvVariantService, CareerProfileService.LoadStructuredForOwnerAsync (public)
- API: CvVariantController (/api/cv), PublicCvController (/api/public-cv)
- Migration AddCvVariants (2 self-contained tables; verified applied on the
running container), 16 tests (resolver/renderer/service), 296 backend green
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 3, version history (list + restore). CareerProfileVersions was already
populated on every save; this makes it usable.
- ICareerProfileService.ListVersionsAsync — the append-only history, newest first,
with the current version flagged.
- RestoreVersionAsync — reapplies a past snapshot NON-DESTRUCTIVELY: it is re-saved
as a new version, so the current state stays in history and the restore is itself
reversible. Syncs the relational children + blob projection like any save.
- Endpoints: GET /career/profile/versions, POST /career/profile/versions/{v}/restore.
Tests (+4): versions listed newest-first with current flagged; restore reapplies
an old snapshot as a new version (history preserved, reversible); restore of a
missing version returns null.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 3, API layer. GET/PUT /api/career/profile — the endpoint the /career editor
uses to read and write the master profile.
- GET: returns the structured profile (assembled from the relational children,
backfilled from the blob if needed) plus a completeness scorecard.
- PUT: validates limits, persists via CareerProfileService (relational children +
append-only version), then serializes the result into
ApplicationUser.ProfileCvStructureJson so the legacy read paths stay in sync.
Identity fields are untouched (they belong to /profile).
- GET /completeness: just the scorecard, for the overview.
- CareerCompleteness: weighted percent + missing sections.
- CareerProfileValidator: item-count/length limits (abuse guard, NOT completeness
— a work-in-progress profile always saves).
Tests (+4): put/get round-trip + projection sync, completeness, over-limit
rejection, empty WIP profile accepted.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 3, service layer. CareerProfileService now maintains the relational children
as the source of truth for structured career data, with the StructuredCvProfile
blob kept as a derived projection.
- SaveVersionAsync additionally syncs the relational children (replace-all,
preserving ItemKeys from the blob item ids; SortOrder = array position) and the
LongTailJson (contact, summary, interests, other sections, metadata).
- New LoadStructuredAsync reads the master profile from the relational children,
lazily backfilling from the ProfileJson blob for profiles that predate Phase 3.
- CareerProfileMapper: the two-way projection between relational rows and
StructuredCvProfile.
Tests (+5): round-trip through relational, item-key preservation, wholesale child
replacement (no orphans), backfill from a pre-Phase-3 blob, empty profile.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Recover the F1 Career Profile foundation + AI-workspace persistence from the
unmerged feature/career-workspace branch, so Phase 2 builds on the documented,
tested target state instead of re-deriving it. Foundation only — CV Builder
commits (variants, ATS badge, rewrite diff) stay deferred per "do not build CV
Builder yet". See docs/career-workspace-branch-assessment.md.
Squashed from 3 branch commits (235e291, 5916f09, 00a035e), resolved against
main + Phase 0:
- CareerProfile + CareerProfileVersion (append-only history), dual-written from
every profile save path via CareerProfileService. ApplicationUser.
ProfileCvStructureJson stays authoritative; the tables mirror it. Stable item
IDs assigned to jobs/education/certifications/projects (the prerequisite for
future variant lineage). CvDateNormalizer for free-text -> YYYY-MM.
- InterviewPrepNote + AiWorkspaceNote: cache AI interview prep / candidate fit /
focus plan keyed by an attachment-context signature, so they stop regenerating
(and re-spending the provider) on every open.
Conflict resolutions (union, favouring current code + Phase 0):
- JobTrackerContext / StartupInitializationExtensions: kept Phase 0's tables and
reconciler blocks, added the career/interview/ai-note tables (both SQLite and
MySQL dialects).
- ProfileCvController: dropped the branch's in-file DTO records (main defines them
in ProfileCvDtos.cs) and the LayoutFamily/AtsRating template fields (deferred
ATS-badge work), keeping main's 7-arg CvTemplateDescriptor.
- JobApplicationsController: kept the branch's cache-check, restored main's
AsNoTracking on the read-only user load.
Tables ship empty (verified dev); nothing to migrate.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Unblocks the documented core workflow and closes the AI-service exposure,
without changing existing behaviour.
Job/JobApplication split (additive; see ADR-002):
- New Job entity (the opportunity) with owner-scoped query filter; nullable
JobApplication.JobId FK. Nothing reads Job yet.
- Migration AddJobEntityAndProspectStages, hand-edited to drop reconciler-owned
tables the scaffolder re-emitted; verified against the real dev DB.
Pipeline: 10 internal stages across three concerns kept separate —
PipelineStage (workflow) / PipelineGroup (UI: NotApplied/Active/Closed) /
PipelineCategory (analytics). Adds Saved/Interested/Preparing/Withdrawn;
keeps Waiting and Ghosted. Kanban shows 3 grouped columns; cards keep a stage
chip and full transitions; drag applies only safe transitions (never infers
Ghosted/Withdrawn).
DateApplied nullable + SavedAt. Cleared when leaving Applied so analytics stay
accurate; the discarded date is preserved as an AppliedDateCleared JobEvent.
AI service lockdown: no host port; private ai_internal network (backend is the
only other member); X-Ai-Service-Token required on all non-/health endpoints;
AI_SERVICE_TOKEN mandatory via compose. Verified backend-only against the live
stack.
Also carries two pre-existing working-tree files (views/ProfilePage.tsx,
views/CareerWorkspacePage.tsx) so the tree is clean for the branch integration.
Tests: +40 backend (247 total), +5 sidecar (16), +15 frontend.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The sessions unit's live smoke test caught the same bug it fixed in
SessionsController also present in TrustedDeviceService and
TwoFactorController's device list: SQLite/Pomelo's EF Core provider
cannot translate DateTimeOffset relational comparisons or ORDER BY to
SQL, so IsDeviceTrustedAsync (the check that skips 2FA for a trusted
browser) and ListTrustedDevices would 500 on real SQLite despite
passing on EF's InMemory test provider. Same fix: equality-only in
the DB query, expiry comparison and sort after materializing.
JWTs were previously fully stateless -- the token alone was the credential
until its own expiry, with no way to list or kill a session server-side. Add
a UserSession table alongside every JWT issued (AppSessionIssuer), embed its
id as a "sid" claim, and check that claim against the DB on every "local"
scheme request (Program.cs OnTokenValidated) so a session can actually be
revoked before its JWT naturally expires. New /api/auth/sessions endpoints
(list, revoke one, revoke-others) plus a Sessions card on the profile page.
Fails closed on a missing "sid" claim: every JWT issued going forward has
one, so a token without it is either pre-deploy (forces one re-login for
already-signed-in users at deploy time, same additive-forward cost the
2FA/trusted-device work on this branch already paid) or forged.
Adds a "trust this device" option to the 2FA challenge: on success, mints a
random token (only its SHA-256 hash is stored), sets it as a new httpOnly,
Secure, SameSite=Strict cookie, and records a TrustedDevice row. AuthController
checks that cookie for the exact signing-in user before gating on 2FA -- a
mismatched user, expired, or revoked device falls through to the normal 2FA
prompt, never errors. TwoFactorController also exposes list/revoke/revoke-all
endpoints for managing trusted devices, scoped to the owning user.
Schema added via the existing raw-SQL reconciler (SQLite + MySQL dialects),
not EF migrations, matching this repo's established pattern.
Adds three layers of account-security hardening, all gated behind the
existing SignInWithAppSessionAsync-equivalent (now AppSessionIssuer) so
every sign-in path -- local, Google, Microsoft -- goes through the same
lockout/2FA checks:
- Per-account lockout: Identity's built-in lockout store (columns already
provisioned, previously unused) is now wired up in AuthController.Login
via IsLockedOutAsync/AccessFailedAsync/ResetAccessFailedCountAsync, 5
failed attempts / 15 min, same generic 401 as wrong-password to avoid
enumeration.
- RFC 6238 TOTP 2FA (Otp.NET) with QR-code setup (QRCoder, fully local/
offline) on a new TwoFactorController: setup requires password
re-confirmation and returns a pending (unconfirmed) secret + QR; the
secret is only persisted as active once verify-setup checks a real
code. Secrets are encrypted at rest via the same IDataProtector pattern
already used for Gmail/Microsoft OAuth refresh tokens.
- Login/OAuth exchange now checks TwoFactorEnabled before issuing a real
session. If enabled, it hands back an opaque, server-side (IMemoryCache)
pending token via a new ITwoFactorPendingTokenService -- deliberately
NOT a JWT, so it can never be presented as a bearer token to bypass the
2FA check on any other endpoint. Only POST /api/auth/2fa/challenge can
redeem it, rate-limited at 5/5min (tighter than password login, since a
6-digit space is far more brute-forceable).
- One-time recovery codes (10 per enable/regenerate, SHA-256-hashed at
rest, shown once in plaintext) accepted in the same challenge endpoint
as an alternative to a TOTP code.
Schema: AspNetUsers gains TotpSecretEncrypted / TotpPendingSecretEncrypted
/ TotpEnabledAtUtc, plus a new TwoFactorRecoveryCodes table, added to both
the SQLite and MySQL dialect blocks in the startup schema reconciler.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- Move inline DTOs to GmailDtos.cs, pure parse helpers to GmailParsing.cs
- Batch per-message existence checks in CreateSuggestedJob/RefreshLinkedThreads
- Remove redundant second pass in RelinkThread, reuse existing HashSet
- Replace ToListAsync+scan with FirstOrDefaultAsync for GmailReviewDecisions lookups
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- Move inline DTOs to JobApplicationDtos.cs, pure static helpers to JobApplicationHelpers.cs
- GetStats aggregates server-side (COUNT/GROUP BY) instead of loading the full table
- Cache RuleSettings via IMemoryCache, keyed per-user (RulesEngine.GetSettings falls back
to per-user UserRuleSettings overrides, so a single global cache key would leak settings
across users)
- Add missing AsNoTracking() to read-only GET endpoints (GetAll, GetById, GetBoard,
GetReminders, GetStatusSuggestion, GetMatchScore, GetCandidateFit, GetFocusPlan,
GetInterviewPrep, GetReadiness)
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Prod was hard-down: InitializeJobTrackerAsync threw an unhandled
MySqlException ("Specified key was too long; max key length is 3072
bytes") while creating IX_JobApplications_OwnerUserId_FollowUpAt,
which crashed Program.Main before the app could start (surfaced to
users as a 500 on Google sign-in, but really affected every request).
Root cause: this reconciler assumes OwnerUserId is varchar(255), but
the live column was provisioned wider by an earlier EF migration,
close enough to the utf8mb4 3072-byte limit that pairing it with a
second column tips a composite index over.
Fix:
- Prefix-index OwnerUserId at 191 chars (safe under the legacy
767-byte-per-column limit, still far wider than the GUID-like
Identity ids actually stored) in every composite/unique index that
includes it, so index creation no longer depends on the column's
actual declared width.
- Wrap each CREATE INDEX in try/catch + LogWarning instead of letting
it propagate: a schema reconciler is best-effort and one failed
index must never crash startup, matching the existing non-fatal
pattern already used a few lines below for legacy-schema ownership
claims.
Backend build + full test suite (177 passing) verified green.
Wave 7. Mirrors the existing Google ID-token-exchange pattern (Program.cs
smart-scheme dispatch, JWT bearer scheme, AuthController exchange/link/
unlink endpoints, ApplicationUser fields, reconciler columns) for
Microsoft Entra ID + personal accounts via the multi-tenant "common"
endpoint.
Google/Microsoft sign-in previously only worked for accounts already
linked to an existing local user -- there was no way to actually sign
up via OAuth. Both exchange endpoints now create a new user when no
match is found and Auth:AllowRegistration is true, same gate as
email/password registration.
Frontend: new MicrosoftAuthCard (MSAL popup flow -- Microsoft has no
vanilla-JS equivalent to Google's Identity Services script) wired into
the login page's provider tabs and the profile page's account-linking
section. REACT_APP_MICROSOFT_CLIENT_ID env var, Auth:MicrosoftClientId
config gate on the backend.
Wave 4 hardening. Wrap untrusted CV/job-description/instruction text
in tools/summarizer prompts with explicit delimiters and an
ignore-embedded-instructions rule, since JD text, recruiter emails,
and free-text candidate background all flow into rewrite/normalize
prompts unescaped today.
Match score previously normalized synonyms (JS/Kubernetes/K8s/etc)
only when scanning the job posting, not when checking the CV corpus,
so a CV using an abbreviation the job spelled out never matched.
SkillTagger.MatchesTag reuses the same synonym regex for both sides.
cesnimda
cesnimda