Move clientIP extraction into a reusable pkg/server/api/clientip
package with RFC 7239 Forwarded header support. Add
pkg/server/jsonutil with helpers for common HTTP error responses
(RenderForbidden, RenderInternalServerError, RenderNotFound,
RenderBadRequest) and use them in the cookie banner handlers.
Signed-off-by: Émile Ré <emile@getprobo.com>
Parse only the first IP from X-Forwarded-For to prevent
the full chain from bypassing anonymization. Add secondary
sort key for deterministic consent selection. Reject origins
with empty port suffix in the validator.
Signed-off-by: Émile Ré <emile@getprobo.com>
Implement config, consent retrieval, and consent recording
endpoints for the JS SDK. IP addresses are anonymized (last
octet zeroed for IPv4, /48 mask for IPv6) before storage.
Signed-off-by: Émile Ré <emile@getprobo.com>
Introduce /cookie-banner/v1/{bannerID}/config endpoint for the JS SDK.
The custom CORS middleware validates each request origin against the
specific banner being requested, preventing cross-customer leakage.
Signed-off-by: Émile Ré <emile@getprobo.com>
Add inherentRiskScore and residualRiskScore to the mutation fragment so
the Relay store gets updated scores. On the backend, use RETURNING to
get the recomputed values from the database generated columns.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Every other LoadAll* method in coredata takes a Scoper parameter
for tenant isolation. LoadAllByCookieBannerID was the only one
that omitted it, making the isolation invariant depend entirely
on callers first loading the banner with a scoped query.
Add scope.SQLFragment to the WHERE clause to match the pattern
used by the paginated sibling LoadByCookieBannerID.
Signed-off-by: Aurélien Sibiril <81782+aureliensibiril@users.noreply.github.com>
- Return 400 instead of panicking on invalid organization_id
- Use generic error message for internal failures
- Drop duplicate validation from initiate handler (kept in tx)
- Make preserveConnectionFields mutate in place
- Remove as type assertions in GoogleWorkspaceConnector
- Use sort.Slice instead of sort.SliceStable
- Consistent error prefixes in Slack sender
Signed-off-by: Aurélien Sibiril <81782+aureliensibiril@users.noreply.github.com>
LoadOneByOrganizationIDAndProvider returns the effective OAuth2
connector for an (organization, provider) pair. When legacy rows
leave multiple candidates behind (pre-auto-reconnect installs), it
picks the row with the widest stored scope set and tiebreaks by
most recent updated_at.
This is correct for Google Workspace where the SCIM bridge's
4-scope row must win over the access review's 2-scope row so the
caller always sees a token that can run SCIM operations. The
selector reads the decrypted scope set through Connection.Scopes,
so it is robust against token-refresh updated_at churn that would
mislead a pure SQL ORDER BY.
Signed-off-by: Aurélien Sibiril <81782+aureliensibiril@users.noreply.github.com>
The title, classification, and document_type columns have been moved to
document_versions. The document_approvers and document_version_approvers
tables have been replaced by approval quorums and decisions.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Add organization_id to cookie_banner_versions, cookie_categories,
and cookie_consent_records so AuthorizationAttributes can return
it directly without JOINing back to cookie_banners. Document the
pattern in contrib/claude/coredata.md.
Signed-off-by: Émile Ré <emile@getprobo.com>
Introduce append-only cookie_banner_versions table with a JSONB
snapshot of consent-relevant configuration (privacy policy URL,
consent mode, expiry, categories and their cookies). Each version
has its own state (DRAFT/PUBLISHED) separate from the banner
lifecycle.
Replace the banner state enum (DRAFT/PUBLISHED/DISABLED) with a
simpler ACTIVE/INACTIVE toggle. Link consent records to the
specific published version the visitor accepted.
Signed-off-by: Émile Ré <emile@getprobo.com>
Now that title lives on document_versions, the full-text search vector
belongs there too. Drop the generated column and GIN index from
documents, recreate them on document_versions, and update the document
filter to query the latest version's search_vector.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
The description field on Document and EmployeeDocument was never
populated. Remove it from the database, GraphQL schema, and Go types.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Title is now owned by document_versions, following the same pattern as
classification and document_type. The documents.title column is made
nullable with a TODO to drop it. Backend loads title from a
latest_versions CTE for ordering purposes only. The frontend resolves
title from the latest version, and DocumentTitleForm now operates on
DocumentVersion using UpdateDocumentVersion mutation.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Replace the per-approver add/remove model with a quorum-based approval
system. Documents now have default approvers that are pre-populated when
requesting approval, and the publish dialog lets users adjust the list
before submitting.
Key changes:
- Add PENDING_APPROVAL document version status with dedicated transitions
- Introduce approval quorums with request/approve/reject/void lifecycle
- Add default approvers per document (stored in document_default_approvers)
with MERGE-based upsert for efficient sync
- Add NoDuplicates validator for slice fields
- Split ALTER TYPE ADD VALUE migrations into separate files (required by
PostgreSQL when run inside transactions)
- Use VOIDED consistently for both quorum status and decision state enums
- Expose void/approve/reject through GraphQL and MCP, with e2e tests
- Add approval management UI: publish dialog with approver selection,
approval list with void support, and external approve/reject page
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Rename the entity across the full stack: database table
(states_of_applicability → statements_of_applicability), Go model,
GraphQL types, MCP specification, CLI commands, frontend components,
routes, and display labels. Includes a migration to rename the table
and its foreign key column.
Widen sidebar from 260px to 280px to fit the longer label.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Introduce the ability to link measures to documents, following the
existing pattern used by controls and risks. This includes:
- Database migration for measures_documents join table
- Coredata MeasureDocument struct with insert/delete operations
- Document service methods for listing/counting by measure ID
- Measure service CreateDocumentMapping/DeleteDocumentMapping methods
- Cleanup of measure-document mappings on document archive
- GraphQL mutations, inputs, payloads, and Measure.documents field
- DocumentConnection.TotalCount support for measure resolver
- MCP linkMeasure/unlinkMeasure updated to support documents
- MCP listMeasureDocuments tool
- Frontend MeasureDocumentsTab with LinkedDocumentsCard integration
- Authorization actions for measure document mapping
- E2e tests for measure document mapping
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Filter out approval decisions where state is APPROVED but
electronic_signature_id is NULL, removing historical approved
entries that lack an electronic signature from the employee
approvals page.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Introduce IN_PROGRESS as a new task state between TODO and DONE across
the full stack: database enum, Go backend, GraphQL, MCP, and frontend.
The task state icon now cycles forward on click (TODO → IN_PROGRESS →
DONE → TODO), and the action dropdown provides explicit "Move to"
options for any state transition. The "All" tab supports drag-and-drop
between state sections to change a task's state.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Follow the same pattern used for classification: document type now lives
exclusively on DocumentVersion. A migration copies existing values from
documents to their versions. The document filter uses a subquery on the
latest version. All three API surfaces (GraphQL, MCP, CLI), resolvers,
frontend, and e2e tests are updated accordingly.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Add go-vcr dependency, dev config for new providers,
connector service changes for access review, connect
schema updates, and unit tests for enum Scan/Value.
Signed-off-by: Aurélien Sibiril <81782+aureliensibiril@users.noreply.github.com>
The existing integer priority field represents positional ordering
within a state, not semantic importance. Rename it to rank and
introduce a new priority field with enum values URGENT, HIGH,
MEDIUM and LOW across the entire stack.
Rank is now scoped to (state, priority) so tasks are ordered
within each priority group. A generated priority_rank column
combines both fields into a single sortable integer for cursor
pagination.
Dragging a task across priority groups updates its priority
automatically based on the drop position neighbors. The backend
first moves the task to the new group then repositions it at the
target rank.
The migration defaults existing rows to MEDIUM priority and
backfills ranks per (state, priority) group.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Introduce two new measure states across the full stack: database
migration, Go coredata, GraphQL schema, MCP specification, and
frontend UI (labels, badge variants, and colors).
Signed-off-by: Bryan Frimin <bryan@getprobo.com>
Introduce dedicated employee-scoped IAM actions and update all
resolvers and frontend mutations accordingly.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Add classification as a filter-only field on documents, resolved from
the latest document version. Expose in GraphQL, MCP, and document list
UI with a selector alongside the document type filter.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Classification now lives exclusively on DocumentVersion. The field is
removed from the Document model, all SQL queries, GraphQL Document
type, SignableDocument type, UpdateDocumentInput, and MCP Document
schema.
New documents still accept classification in CreateDocumentInput,
applied to the first version. New drafts inherit classification from
the previous version. PDF generation uses the version classification.
The drawer allows editing classification on draft versions via the
updateDocumentVersion mutation. Classification is read-only on
published versions.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Add ISC license headers to all .go, .ts, .tsx, and .sql files
using each file's git history to determine the correct copyright
year or year range. Trademarked icons (brand logos, vendor logos,
compliance framework logos) are excluded.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Introduce a two-part version scheme (major.minor) for documents.
Drafts start at 0.1 and increment minor on each new draft. Publishing
as minor keeps the current version, publishing as major bumps to the
next major.0. Both current_published_major and current_published_minor
are tracked on the document for exact version lookups.
Signatures and approval quorums aggregate across all versions sharing
the same major number using CTE joins. Approval page mutations spread
the decision fragment so Relay updates the version row state without
requiring a page refresh.
GraphQL, MCP, and service layer expose separate publishMajor and
publishMinor mutations instead of a single mutation with a type enum.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Introduce a complete approval system for document publishing. Document
versions can now require approval from selected reviewers before being
published, with automatic publishing once all approvers have approved.
- Add approval quorum and decision tables with backfill migration
- Implement request approval, approve, and reject flows with electronic
signature support for approve decisions
- Add employee approvals page with dedicated tab and pending approvals view
- Add changelog field to publish and request approval flows
- Pre-select previous version's approvers in the publish dialog
- Show quorum approvers in document list with 100 approver hard limit
- Expose approval workflow through GraphQL, MCP, and CLI
- Remove legacy default approvers feature entirely
- Add comprehensive e2e test coverage for approval workflows
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Introduce a rank-style priority on tasks, scoped by
(organization_id, state). New tasks auto-assign the next
priority. Reordering uses the same CTE-based algorithm as
trust center references and compliance external URLs.
Exposed through GraphQL, MCP, and the PRIORITY order field.
Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>
Previously the worker reset failed descriptions back to PENDING,
causing infinite retry loops for permanently broken evidence.
Add a FAILED status so these items stop cycling and can be
investigated.
Signed-off-by: Bryan Frimin <bryan@getprobo.com>
Introduce a background worker that automatically generates
compliance-focused descriptions for uploaded evidence files
using configurable LLM providers. Descriptions are surfaced
across all interfaces: GraphQL API, MCP API, CLI, and the
console UI.
Key changes:
- Multi-provider LLM config with per-agent settings (pointer
types for Temperature/MaxTokens to preserve zero values)
- Evidence description worker with bounded concurrency
- EvidenceDescriptionStatus typed enum with PostgreSQL enum type
- New `prb evidence` CLI commands (list, view, delete)
- Evidence description displayed in console table and preview
- Migration only marks evidences without files as completed
Signed-off-by: Bryan Frimin <bryan@getprobo.com>
Trust centers were created without setting SearchEngineIndexing,
defaulting to an empty string in the database. This caused scan
errors when loading trust centers.
Add a migration to fix corrupted rows, set a DEFAULT on the
column, and add a CHECK constraint. Also set the field explicitly
when creating new trust centers.
Signed-off-by: Bryan Frimin <bryan@getprobo.com>
Introduce dataloadgen-based dataloaders to batch individual
record-by-ID fetches in GraphQL resolvers into single SQL
queries. Each entity type (organization, framework, control,
vendor, document, risk, measure, task, file, report, profile)
gets a LoadByIDs method in coredata and a GetByIDs service
method with variadic arguments and dedicated collection return
types. Resolvers now use dataloader.FromContext instead of
direct service calls for single-record lookups.
Signed-off-by: Bryan Frimin <bryan@getprobo.com>
Add search engine indexing toggle, robots.txt, and sitemap.xml
generation for compliance pages. Replace checkboxes with toggle
components in the compliance page UI and add an "Open" button
in the page header to quickly access the live compliance page.
The search engine indexing toggle is disabled when the compliance
page is inactive.
Signed-off-by: Bryan Frimin <bryan@getprobo.com>
