BigBrain v1.0.1

Fixed

• Restored page writes for registered legacy Brains that do not yet have a
  page-schemas.yaml registry. Owning MCP services now apply an atomic,
  versioned, additive compatibility migration before serving requests. The
  migration derives permissive categories from each Brain's schema_dirs,
  preserves Brain-local definitions, and never rewrites invalid or newer
  registries.

Agent update actions

• Restart each owning MCP service after installing this update. On startup the
  service creates or advances the Brain-local page schema registry when safe.
  Do not copy a registry between Brains. Review schemas/list, then verify one
  normal page create and read-back through that Brain's MCP.
• Source installations should update to v1.0.1, run npm install and
  npm link, then restart only source-managed local services. Server-managed
  deployments should promote the immutable v1.0.1 image and verify /ready
  plus authenticated tools/list after rollout.

Verification

• Focused page schema, page operation, hosted runtime, and MCP tests pass,
  including missing-registry writes by a non-admin creator, additive custom
  categories, idempotence, invalid-registry rejection, and future-version
  no-rewrite coverage.
• Full test suite passed: 684 tests passed, 7 PostgreSQL integration tests were
  skipped because TEST_DATABASE_URL was not configured, and 0 failed.
• Dashboard production build and standalone MCP bundle verification passed.

BigBrain v1.0.0

Added

• Added a Brain-local page schema registry with MCP tools to inspect,
  bootstrap, register, update, and delete schema definitions, plus health
  checks for schema compliance and first-class page summaries.
• Added first-class frameworks, articles, and videos collections,
  multi-image page attachments, article capture, and dedicated page, search,
  Substack, podcast, YouTube, email-management, and ingestion skills.
• Added the Stream dashboard with source and collection filters, attached-image
  browsing, complete source counts, and bounded infinite pagination.
• Added Comprehension and Keep in Touch playbooks with private overlay state,
  bounded reviews, resumable cursors, and dashboard surfaces.
• Added one shared verified RSS and Substack ingestion ledger with atomic
  claims, retryable outcomes, explicit partial captures, scoped read-back
  evidence, and per-feed retained-coverage bounds.

Changed

• Search now embeds and returns complete ranked page evidence for harness-side
  synthesis, overlaps lexical and semantic retrieval, and applies bounded
  reranking to complete page content.
• New structured timeline entries require source provenance. Task and page
  mutations now share the canonical page writer and reject storage-oriented
  timeline prose.
• Canonical pages now support a first-class summary stored in frontmatter and
  rendered separately from their detailed body. Health reports missing
  summaries as maintenance findings.
• Background event workers now recover orphaned claims and queue backpressure,
  isolate native computer use, and expose only the MCP and browser capabilities
  required by ingestion work.
• Focused graph navigation, labels, source flow, node controls, lifecycle
  handling, and page reading views have been substantially refined.

Fixed

• Fixed hosted browser OAuth onboarding and remote dashboard capability
  handling.
• Fixed graph focus after rotation, transient particle teardown, label overlap,
  and focused-network recovery.
• Fixed RSS retry visibility, verified completion checks, and Granola route
  verification so queue or worker prose cannot falsely prove ingestion.

Removed

• Removed the MCP natural-language query answer tool. Use search for
  complete evidence and let the calling harness synthesize the answer. The CLI
  query compatibility command remains retrieval-only.
• Removed the standalone conversation-ingest skill. Conversation capture is now
  part of the general ingestion workflow.

Agent update actions

• This is a major upgrade. Before updating, back up the Brain directory,
  runtime database, ~/.config/bigbrain, and the machine-local RSS/Substack
  ledger. Existing schema-v1 databases are migrated to embedding schema v2 on
  first use; rollback to an older runtime is not supported after migration.
• Source installs: update to v1.0.0, run npm install, npm link, and
  npm run build:dashboard, then restart the desktop, dashboard, MCP, event
  ingestor, and RSS services.
• Existing Brain configs do not automatically gain new schema directories.
  Merge frameworks, articles, and videos into schema_dirs without
  overwriting local configuration. Bootstrap the page schema registry, review
  it, and commit page-schemas.yaml before enforcing custom schemas.
• Run a full bigbrain sync --json with the deployment's embedding credentials
  configured so complete-page embeddings are generated. Then run
  bigbrain health --json and resolve missing page summaries, strict-schema
  findings, and legacy task fields. Only exact blank legacy assignees or
  source strings may be normalized automatically to empty arrays; review all
  other values manually.
• Refresh active bundled skills and remove any active
  bigbrain-conversation-ingest copy. Review the new email-ingest,
  Keep in Touch, and Substack-reconciliation automations before activating
  them, and preserve local schedules, targets, and customizations.
• Server and container deployments must persist the Brain, database, OAuth
  state, and the shared ledger path configured by
  BIGBRAIN_SUBSTACK_LEDGER_PATH. Never reconstruct verified article outcomes
  from feed cursors or prose logs.
• MCP clients must replace query calls with search. Structured timeline
  writers must provide complete provenance with event_id, source_type,
  source_label, and source_message for every new entry.
• Verify /live, /ready, the reported runtime version, authenticated
  tools/list, search, page schema tools, RSS/Substack ledger status,
  Comprehension and Keep in Touch playbooks, bigbrain sync --json,
  bigbrain health --json, and npm test after deployment.

Verification

• Local-data compatibility audit passed with the required major-version and
  migration actions above. Legacy timeline entries remain readable, task enum
  values are unchanged, and shared ledger migrations preserve existing rows.
• Full test suite passed: 678 tests passed, 7 skipped because
  TEST_DATABASE_URL is not configured, and 0 failed.
• Dashboard production build and standalone MCP bundle verification passed as
  release gates before tagging.

BigBrain v0.28.0

[0.28.0] - 2026-09-03

Added

• Added browser-first OAuth entry and retry routes so /connect, dashboard
  redirects, and protected page links can start Google sign-in without the
  BigBrain CLI.
• Added hosted OAuth owner recognition. An active owner member is a
  server-side superuser above admin and can use all MCP tools, including
  member and custom-role administration.

Changed

• OAuth error pages now preserve the requested dashboard or page route and
  offer a browser retry with another Google account.
• Hosted MCP clients continue to use standard OAuth discovery and scoped
  tokens. Ordinary members remain least-privilege; admin still requires
  brain:admin, while an active owner is elevated by the Brain role.

Agent update actions

• Source and standalone MCP installs: update to the v0.28.0 tag, run
  npm install, npm link, and npm run build:dashboard, then restart the
  BigBrain desktop, dashboard, and source-managed MCP services. Verify
  /connect offers browser Google sign-in, protected page links redirect to
  OAuth, MCP protected-resource metadata advertises the service, and
  bigbrain health --json plus npm test pass.
• Hosted deployments: deploy the v0.28.0 BigBrain runtime while preserving
  the Brain directory, database, OAuth token store, and configured email
  allowlist. Verify /live, /ready, browser dashboard sign-in, and
  authenticated MCP tools/list. Confirm the intended owner is an active
  owner member before relying on owner superuser access.
• No Brain page, task, member, database, or token migration is required.
  Existing OAuth tokens remain valid. The owner elevation is determined from
  the active member role on each MCP request, so no token regeneration is
  needed for existing owners.

Verification

• Local-data compatibility audit passed. The release adds only additive OAuth
  routes, error-page behavior, and role-aware authorization; it does not alter
  persisted page, task, member, or database schemas or narrow existing values.
• Full test suite passed: 501 tests passed, 7 skipped because
  TEST_DATABASE_URL is not configured, and 0 failed.
• Dashboard production build and standalone MCP bundle verification passed as
  release gates before tagging.

BigBrain v0.27.1

Full Changelog: v0.27.0...v0.27.1

BigBrain v0.27.0

Added

• Added canonical components, products, and processes page collections,
  filing rules, and folder recommendations for infrastructure knowledge.
• Added focused graph network and timeline views with provenance-aware timeline
  updates.

Changed

• Added optional per-feed RSS guidance so each registered feed can define what
  to prioritize, what to ignore, and how its article events should be handled.
• Tightened the source-article relevance gate so low-value articles can be
  ignored before creating a raw artifact or Brain page.
• Preserved the RSS listener description as source identity while using the
  dedicated guidance field for ingestion scope, with legacy descriptions kept
  as a compatibility fallback.

Agent update actions

• Source installs: update to the v0.27.0 tag, run npm install, npm link,
  and npm run build:dashboard, then restart the BigBrain desktop, dashboard,
  and source-managed MCP services. Verify bigbrain sync --json,
  bigbrain health --json, MCP initialization, and npm test.
• Review registered RSS listeners and move any ingestion instructions that are
  currently in description into the new guidance field. Keep description
  for the feed's identity and purpose. Existing feeds remain compatible because
  RSS descriptions are used as a fallback when guidance is absent.
• No Brain page, timeline, or database migration is required. The new
  components, products, and processes collections are created by the
  normal Brain layout setup when needed.
• Active Codex skill copies remain machine-local. Refresh the packaged source
  article-ingestion skill from the release and restart event-ingestor services
  so the relevance gate and per-feed guidance are active.

Verification

• Full test suite passed.
• Dashboard production build passed.
• Standalone MCP bundle build passed, including its release manifest and
  checksum.
• Package dry-run and source version, lockfile, tag, and artifact metadata
  checks passed.

BigBrain v0.26.0

Added

• Added optional Brain knowledge domains, with stable IDs, human-friendly names,
  and authoritative usage guidance stored in a Brain-specific domains.yaml
  registry.
• Added explicit MCP domain registry CRUD through domains/list,
  domains/get, domains/create, domains/update, and domains/delete.
• Added domain membership to page create, raw-file-with-page, and page update
  operations, while rejecting unregistered IDs and never auto-creating domains.
• Added dashboard graph domain filtering and domain-aware page previews.

Changed

• Made page tool schemas dynamically project the current domain registry through
  Zod, including each domain's guidance in the field description.
• Added health validation for invalid registries, unregistered memberships, and
  duplicate memberships.
• Made domain-scoped previews redact links to pages outside the selected domain
  at render time while preserving canonical Markdown unchanged.

Agent update actions

• Source installs: update to the v0.26.0 tag, run npm install, npm link,
  and npm run build:dashboard, then restart the BigBrain desktop, dashboard,
  and source-managed MCP services. Verify bigbrain sync --json,
  bigbrain health --json, MCP initialization, and npm test.
• No migration is required for existing pages, tasks, databases, or services.
  The domains page field is optional, and pages without it remain valid.
  Preserve and back up any existing Brain domains.yaml file with the Brain.
• If pages already contain domains memberships, ensure each ID is explicitly
  present in that Brain's registry or remove the membership, then run
  bigbrain sync --json and bigbrain health --json until the domain findings
  are clear. Do not invent registry entries during routine ingestion.
• Use domains/create, domains/update, and domains/delete only when the
  user explicitly curates the domain taxonomy. Use page tools only to assign
  already-registered IDs; the dynamic tools/list schema is the current source
  of available memberships and their guidance.
• Packaged desktop and standalone MCP installs should update together when
  practical, preserve Brain and database volumes, and verify the dashboard
  domain filter plus /health and /ready after restart.

Verification

• Serialized full test suite passed: 497 tests passed, 6 skipped because
  TEST_DATABASE_URL is not configured, and 0 failed.
• Dashboard production build passed.
• Local dashboard development smoke test passed with synthetic infrastructure,
  startup, and multi-domain meeting pages. Domain filtering returned the
  expected memberships, and scoped previews preserved in-domain links while
  rendering out-of-domain links as labels.

BigBrain v0.25.2

[0.25.2] - 2026-09-02

Fixed

• Fixed the MCP release workflow to initialize Docker Buildx before exporting
  its multi-architecture image cache and publishing the server image.

Agent update actions

• Update to the v0.25.2 tag and repeat the 0.25.1 verification actions.
  No additional Brain data or runtime migration is required.

Verification

• Local tests, dashboard build, MCP bundle build, package dry run, and
  universal macOS packaging checks pass. The release workflows must complete
  successfully from v0.25.2 before this release is reported as published.

BigBrain v0.25.1

Full Changelog: v0.25.0...v0.25.1

BigBrain v0.24.0

See CHANGELOG.md for complete release notes. Highlights: ~/.config/bigbrain/brains.json is now the canonical connected-brain catalog for CLI and desktop; legacy macOS desktop registry migration is recoverable via registry.json.legacy; desktop dropdown and force-graph startup race are fixed.

BigBrain v0.23.0

Added

• Added a guided desktop brain flow for creating a private local Brain,
  connecting an existing local Brain folder, or connecting to an existing
  BigBrain service.
• Added Brain name and description capture, local Brain discovery, explicit
  GitHub-backup preference, and secure API-key selection during setup.
• Added actionable inbound-event draining with bigbrain events drain, plus
  clearer health and maintenance guidance for failed or quarantined events.
• Added bounded local dashboard error reports that save diagnostics locally
  without sending them automatically.
• Added exact YouTube metadata, stable-source identity, chapter preservation,
  and pre-write validation for media ingest.
• Added animated graph membership transitions for timeline scrubbing, live
  graph updates, and chronological initial graph reveal.

Changed

• Made local desktop setup and cancellation return to the dashboard cleanly,
  with readable handling for missing installers, occupied ports, and services
  that do not become ready.
• Separated the developer desktop identity and launch ownership from the
  release desktop and shared local MCP services, while keeping desktop event
  configuration in the shared user configuration directory.
• Added safe BIGBRAIN_CONFIG_DIR overrides while preserving the existing
  ~/.config/bigbrain default for event registry, inbox, and ingestor state.
• Made RSS filing reject paths outside established Brain collections and made
  source-article ingest choose relevance before writing source artifacts.

Fixed

• Fixed the add-Brain route that previously rendered the unusable blank
  “Choose a brain, or add another one” page.
• Fixed setup-flow escape behavior and prevented duplicate developer desktop
  launchers from competing for the same port.
• Fixed graph membership updates appearing instantaneous by retaining entering
  and exiting nodes and links while their visual state transitions complete.

Agent update actions

• Packaged desktop installs: update to BigBrain v0.23.0 and restart the app.
  Use the client update control when available, then verify that the desktop
  and each desktop-owned local MCP service report 0.23.0. Existing source,
  remote, newer, or unknown-owner services must remain untouched.
• Source installs: update to the v0.23.0 tag, run npm install, npm link,
  and npm run build:dashboard, then restart every BigBrain desktop, dashboard,
  and local MCP service. Verify with bigbrain --version,
  bigbrain sync --json, bigbrain health --json, npm test, and
  npm pack --dry-run.
• If using the new local setup flow, choose the GitHub-backup preference that
  matches the Brain owner’s intent. This preference applies to the new or
  adopted Brain registration and does not rewrite existing Brain content.
• No Brain Markdown migration, Brain profile schema migration, task/page field
  migration, database migration, member-role migration, OAuth/API-key
  migration, or event-registry migration is required. Existing task values for
  status, readiness, priority, assignees, source, and execution mode
  remain valid. Existing event files continue to use their current default
  location unless BIGBRAIN_CONFIG_DIR is deliberately configured.
• Server deployments: deploy only the immutable v0.23.0 release tag, preserve
  Brain and database volumes, and verify /live, /ready, authenticated MCP
  tool listing, and owning-MCP read-back after restart.

Verification

• Full npm test suite: 457 passed, 6 skipped, 0 failed.
• Dashboard production build completed successfully.
• Graph transition, local setup, service ownership, error reporting, event
  handling, YouTube provenance, and skill-template regressions passed.
• Local-data compatibility audit completed against v0.22.0; no persisted
  Brain or task migration is required.
• Developer desktop restarted and verified as the single active Dev app on
  its expected local port.