Canonical Timeline User Guide

Welcome to the Canonical Timeline user guide. This documentation covers everything you need to work with investigations, timelines, events, claims, sources, entities, and relationships.


Quick Navigation

SectionDescription
Core ConceptsThe six-layer data model: investigations, timelines, events, claims, sources, entities
Working with InvestigationsCreating investigations, navigating lenses (Overview, Timeline, Evidence, Knowledge, Geography, Analytics, Background)
WorkspaceThe investigation index — browse, filter, star, and create investigations
Working with TimelinesCreating timelines, browsing events, filtering, and exporting within investigations
Working with EventsCreating events, setting time precision, linking entities, and keyboard navigation
Claims and SourcesCreating claims with source attribution, confidence scores, and source management
Entities and RelationshipsCreating entities, managing relationships, entity graphs, and entity reconciliation
Keyboard ShortcutsComplete shortcut reference — global, page, and component scopes
File UploadsUploading files to sources, checksum deduplication, and quota management
Search and ExportFull-text search, CSV/JSON/Markdown export, and entity graph PNG export
Revision HistoryReading the audit log, before/after snapshots, and soft-delete
Account and SettingsAccount management, members, invitations, API keys, subscription tiers, and the Developer API page

Getting Started

This section walks you through your first session — from login to a completed investigation.

1. Create Your Account

Navigate to /register to create an account. You will need:

  • A display name (visible to other members of your account)
  • An email address
  • A password

After registration, check your email for a verification code. Enter it at /verify to activate your account.

If you were invited by an existing account, use the invitation link provided to you. Invitation-based registration auto-assigns you to that account with the role specified by the inviter.

2. Your First Login

After verification, log in at /login. You will land on the workspace page (the investigation index), which shows a list of investigations in your account.

The sidebar is divided into five sections:

Workspace

  • Investigations — The investigation index (your home page)
  • Recent — Recently viewed investigations
  • Starred — Your starred investigations

Tools

  • Review Queue — Gaps and items needing attention across all investigations
  • Search — Full-text search across all data

Management

  • Entities — Named actors (people, organizations, etc.)
  • Sources — Reference documents
  • Type Management — Custom type taxonomies for your account
  • Entity Reconciliation — Find and merge duplicate entities

Account

  • Account Management — Manage members, invitations, and subscription
  • Settings — Profile, API keys, and personal settings

Documentation

  • Documentation — This user guide
  • Developer API — MCP configuration and REST API reference (requires Professional or Enterprise tier)

If your account has demo datasets enabled, you will see pre-populated investigations like "Apollo 11 Mission," "Watergate Investigation," or "Operation Ghost Courier." These are read-only samples that let you explore the application before creating your own data.

3. Create Your First Investigation

An investigation is the top-level container — it holds a timeline, events, claims, sources, entities, and relationships. To create one:

  1. From the workspace page, click the + New Investigation button.
  2. In the dialog:

- Title: Give your investigation a descriptive name (e.g., "Acme Corp Investigation").
- Status: Choose the current phase:

- draft — Planning and initial setup

- active — Investigation is in progress

- paused — Temporarily on hold

- resolved — Investigation is complete

- archived — Retired for reference

- Summary: An optional description of what this investigation covers.

  1. Click Create.

You will be taken to the investigation detail page, which opens to the Overview lens.

4. Explore the Investigation Lenses

Each investigation has seven lenses, accessible via tabs below the header:

LensPurpose
OverviewHealth scores, metric cards, recent activity, key entities, top sources
TimelineFull event sequence with filters, search, confidence slider, and inline editing
EvidenceClaims and sources for this investigation — filter, review, and manage
KnowledgeEntities and relationships — explore the entity graph and connection network
GeographyInteractive map showing event locations across the investigation
AnalyticsHealth scores, gaps (unsourced events, orphan entities), and trends over time
BackgroundNotes, open questions, and resolved questions for the investigation

5. Add Your First Event

From the Timeline lens:

  1. Click the + Event button in the toolbar, or press N.
  2. In the dialog:

- Title: A concise name for the event (e.g., "Board meeting at HQ").
- Canonical Time: The date and time of the event. You can be specific or approximate.

- Time Precision: How certain you are about the time:

- exact — You know the precise time

- estimated_date_time — Approximate but you have a specific estimate

- exact_time_period_range — Known range with certain bounds

- estimated_time_period_range — Estimated range

- Description: Context and details about the event.

- Status: The verification status (e.g., reported, confirmed, alleged).

  1. Click Create.

The event appears as a card in the timeline sequence.

6. Add a Claim to Your Event

Claims are factual assertions attributed to sources. Every claim must cite a source.

  1. Expand an event card by clicking it, or press Enter when it is selected.
  2. Click + Add Claim in the context panel.
  3. In the dialog:

- Claim Text: The factual assertion (e.g., "The CEO was present at the board meeting").
- Source: Search for and select an existing source, or create a new one inline.

- Status: The current verification grade:

- rumored — Unverified, heard through informal channels

- reported — Reported but not yet assessed

- unverifiable — Cannot be verified

- verifiable — Can be verified based on available evidence

- fact — Verified as accurate

- Confidence: A score from 0 to 100 indicating how confident you are in this claim.

  1. Click Create.

7. Create a Source

If you need to create a source before adding a claim:

  1. Navigate to Sources in the sidebar (under Management).
  2. Click + New Source.
  3. In the dialog:

- Title: The source document's name.
- Source Type: Choose from 175+ types (e.g., news_article, court_document, intelligence_report).

- Author: Who created or published the source.

- Publication Date: When it was published.

- URL: A link to the source if it exists online.

  1. Click Create.

You can return to the event and assign claims to this source.

Entities represent the people, organizations, and objects involved in your investigation.

  1. From an event, click + Link under the "Linked Entities" section.
  2. Search for an existing entity or create a new one:

- Canonical Name: The full, official name (e.g., "John D. Rockefeller").
- Aliases: Alternative names they are known by.

- Entity Type: Choose from 170+ types (e.g., person, corporation, government_agency).

  1. Click Link.

Entities can also be linked to each other via relationships (e.g., "works for," "owns," "reports to").
From the Knowledge lens, explore the entity graph to see the full relationship network.

9. Review Your Work

Navigate to the Review Queue (sidebar > Tools > Review Queue). The review queue surfaces gaps across all investigations:

  • Events Without Claims — Events that lack evidence
  • Low Confidence Claims — Claims with confidence below 50%
  • Orphan Entities — Entities not linked to any event
  • Pending Status Review — Claims that may need their status updated

Work through these queues to ensure every event has sourced claims and every claim has been evaluated.

For investigation-specific analytics, use the Analytics lens to view health scores, gaps, and trends scoped to a single investigation.

10. Export Your Investigation

From the Timeline lens, click CSV, JSON, or Markdown in the toolbar to download a structured export. These exports include events, claims, sources, and entity references — suitable for sharing, archiving, or feeding into other tools.


Conventions in This Guide

  • Bold text indicates UI elements: buttons, labels, field names, dialog titles.
  • <kbd> tags indicate keyboard keys and shortcuts.
  • > Blockquotes indicate important notes or warnings.
  • Screenshots are referenced by filename. These live in public/screenshots/ and may be referred to as screenshot-name.png throughout the guide.
  • Paths like "sidebar > Management > Entities" mean: click the Entities link in the left sidebar, under the Management section.

Next: Core Concepts →