Workspace / PRODUCT DOSSIER

Nimbalyst

A visual workspace for agents, plans, and the things you build.

Desktop & iOS · Open source · MIT · desktop & iOSRevision 2026-09-13.4 · unreviewedNeeds recheck · recheck 2026-10-130 attestations · checked 2026-09-13

What we do not know yet · 3 open gaps

  • Measure how agents coordinate when tasks touch the same files.
  • Test recovery and session continuity across desktop and mobile.
  • Inventory leads for native cross-session messaging, live context meters and internal orchestration APIs still need matching public source evidence. Claude Agent team behavior must be checked against the installed runtime version.

Applies to: Unpinned public documentation as of 2026-09-13 · Documentation review only

Applicability, 3 recheck triggers & verification plan

Applicability

Unpinned public documentation as of 2026-09-13

Documentation review only

Excludes: Runtime behavior; Performance comparisons; Security guarantees

Recheck when

  • Documentation changes
  • New product release
  • License change

Needs recheck · due

Verification plan

  1. Pin a product release or repository commit and record the host environment.
  2. Create a disposable repository with a passing baseline test.
  3. Run a bounded change with two agents where supported; record permissions and logs.
  4. Record a failed attempt, actual outcome, and artifacts without secrets.
  5. Have another person reproduce the result before labeling it reproduced.

Provenance & review

Testedknowhow initial research

Commissioned by a Nimbalyst maintainer; not an independent comparison.

Revision 2026-09-13.4 · unreviewed · checked

0 attestations · Content: Unlicensed prototype; redistribution decision pending

Capability claims · 41 / 46 sourced

Standing condition: these are documentation claims at the source capture date, with outcome “claimed”; no runtime execution is recorded. Dossier applicability and limitations apply throughout. Expand a row for its statement, specific scope and contradictions. Numbered citations resolve to the source locator and capture date below. Missing evidence means Unknown, not No. Sources are unpinned unless marked “Pinned”; this marks source stability, not stronger support or runtime verification. Reuse counts refer to claims in the full dossier; repeated citations do not represent independent sources.

Yes Documented within scope Partial Limited form of the feature No Explicit evidence of absence Unknown No evidence recorded. These are support assessments, not quality scores or runtime verification.

Orchestration

Parallel agentsYes
Explore Parallel agents →

Runs parallel coding sessions organized on a board.

Git worktreesYes
Explore Git worktrees →

Creates separate Git worktree sessions when requested.

Agent choiceYes
Explore Agent choice →

Runs Claude Code and Codex as first-class providers.

Task managementYes
Explore Task management →

Tracks plans, issues and decisions on boards.

AutomationsYes
Explore Automations →

Runs recurring AI tasks from scheduled Markdown documents.

Scope: Configured automation schedules; not proof of unattended execution while the app is closed.

Attention notificationsYes
Explore Attention notifications →

Notifies when parallel work completes.

Agent coordination

Agent delegationYes
Explore Agent delegation →

Claude Agent can spawn parallel teammates.

Scope: Experimental Agent Teams setting; behavior depends on the configured Claude runtime.

Workstream groupingYes
Explore Workstream grouping →

Groups related sessions with shared tabs and an edited-files view.

Scope: Grouping does not provide Git isolation.

Session lifecycle boardYes
Explore Session lifecycle board →

Organizes sessions into five lifecycle phases.

Scope: Desktop Session Kanban.

Session continuity

Session resumeYes
Explore Session resume →

Searches, resumes and branches saved sessions.

Session branchingYes
Explore Session branching →

Branches a session while retaining the original conversation.

Scope: Session management surface.

Prompt queuesYes
Explore Prompt queues →

Busy sessions accept queued prompts from mobile.

Scope: iOS companion; offline prompts transmit after reconnection.

Restart recoveryPartial
Explore Restart recovery →

Scheduled session wakeups survive app restarts.

Scope: Pending wakeups only; the app must be running to fire. Not arbitrary process recovery.

Session import & exportYes
Explore Session import & export →

Imports existing Claude Code CLI conversations.

Scope: Import direction; eligible local CLI sessions. Export not established by this source.

Observability

Usage & cost reportingYes
Explore Usage & cost reporting →

Reports historical session, token, project and tool usage.

Scope: Tool counts are not per-tool token consumption; monetary cost is not established here.

Review & delivery

Change reviewYes
Explore Change review →

Shows tracked text edits as reviewable diffs in supported editors.

Git operationsYes
Explore Git operations →

Stages changes and manages branches visually.

Pull request workflowYes
Explore Pull request workflow →

Reviews pull requests and opens them in worktrees.

Checkpoints & rollbackYes
Explore Checkpoints & rollback →

Restores earlier document versions from file history.

Scope: File history, not complete agent/process rollback.

Selective change approvalYes
Explore Selective change approval →

Keeps or reverts edits in individual worktree files.

Scope: File-level review before committing; not a claim of hunk-level controls.

Merge & conflict assistanceYes
Explore Merge & conflict assistance →

Merges worktree branches and auto-stashes during rebase.

Scope: Git repository and Developer Mode; conflict-resolution behavior still needs verification.

Work surface

Visual authoringYes
Explore Visual authoring →

Provides visual editors for markdown, diagrams, mockups and data models.

Code editingYes
Explore Code editing →

Includes Monaco source-code editing.

Terminal accessYes
Explore Terminal access →

Provides an integrated terminal.

Browser interactionYes
Explore Browser interaction →

Provides Chromium navigation and agent click, type and screenshot tools.

Scope: In-app browser/editor surface.

Rich document authoringYes
Explore Rich document authoring →

Provides editable WYSIWYG Markdown documents.

Scope: Native document editor.

Visual selection contextYes
Explore Visual selection context →

Sends selected mockup elements and drawn annotations to the agent.

Scope: Mockup editor supplies rendered context and HTML.

Voice interactionYes
Explore Voice interaction →

Supports spoken interaction and voice-triggered coding sessions.

Scope: Alpha; desktop and iOS. Includes spoken replies and supported approval actions.

Context & models

MCP toolsYes
Explore MCP tools →

Connects agents to MCP servers.

Project instructionsYes
Explore Project instructions →

Uses project CLAUDE.md and AGENTS.md instructions.

Skills & custom agentsYes
Explore Skills & custom agents →

Supports reusable slash commands and skills.

Model choiceYes
Explore Model choice →

Supports multiple providers and user-supplied API keys.

Local modelsYes
Explore Local models →

Documents local inference through LM Studio.

Project memory & retrievalYes
Explore Project memory & retrieval →

Memory extension indexes project knowledge and recalls durable facts.

Scope: Requires Nimbalyst Memory extension; local derived index.

Extension SDKYes
Explore Extension SDK →

Third-party packages can register editors with structured agent hooks.

Scope: Extension EditorHost contract; extension installation required.

User-owned credentialsYes
Explore User-owned credentials →

Connects user-provided API keys or existing agent subscriptions.

Scope: Provider-specific credentials; recurring model charges are separate.

Execution & access

Mobile accessYes
Explore Mobile access →

Offers a mobile app for session management.

Team collaborationYes
Explore Team collaboration →

Provides shared documents and team trackers.

Tool permissionsYes
Explore Tool permissions →

Configures tools and filesystem permissions per project.

Remote agent controlYes
Explore Remote agent control →

Starts and redirects desktop sessions and answers approvals from iOS.

Scope: Desktop must be connected; code diff review remains on desktop.

Team roles & accessYes
Explore Team roles & access →

Owners and admins manage member roles and access.

Scope: Nimbalyst Teams; Members & Roles is labeled beta.

Unknown · 5 capabilities without evidence

Interface images & local attachments

Sources · 17

[1]Nimbalyst official documentation ↗5 / 41 claims
vendor-documentation · locator & rights

Public product overview; desktop and iOS licensing scope. Hosted services are not covered by that license statement.

Link and authored summary only; upstream content retains its own rights.

[2]Nimbalyst repository ↗0 / 41 claims
repository · locator & rights

Repository overview and license; no product execution.

Upstream license applies.

[3]Nimbalyst · features ↗19 / 41 claims
vendor-documentation · locator & rights

Agent Sessions; Developer Features; Planning; Collaboration; Agent Integration; Open & Secure; Mobile.

Authored summary and link only; upstream rights retained.

vendor-documentation · locator & rights

Named feature sections in agent teams and super loops.

Authored summary and link only; upstream rights retained.

[5]Nimbalyst · workstreams ↗1 / 41 claims
vendor-documentation · locator & rights

Named feature sections in workstreams.

Authored summary and link only; upstream rights retained.

[7]Nimbalyst · session features ↗1 / 41 claims
vendor-documentation · locator & rights

Branch and archive.

Authored summary and link only; upstream rights retained.

[8]Nimbalyst · mobile app ↗2 / 41 claims
vendor-documentation · locator & rights

Named feature sections in mobile app.

Authored summary and link only; upstream rights retained.

vendor-documentation · locator & rights

Named feature sections in import claude code sessions.

Authored summary and link only; upstream rights retained.

[10]Nimbalyst · ai usage report ↗1 / 41 claims
vendor-documentation · locator & rights

Named feature sections in ai usage report.

Authored summary and link only; upstream rights retained.

[11]Nimbalyst · worktrees ↗2 / 41 claims
vendor-documentation · locator & rights

Named feature sections in worktrees.

Authored summary and link only; upstream rights retained.

[12]Nimbalyst · mockups ↗1 / 41 claims
vendor-documentation · locator & rights

Named feature sections in mockups.

Authored summary and link only; upstream rights retained.

[13]Nimbalyst · voice mode ↗1 / 41 claims
vendor-documentation · locator & rights

Named feature sections in voice mode.

Authored summary and link only; upstream rights retained.

[14]Nimbalyst · setup teams and orgs ↗1 / 41 claims
vendor-documentation · locator & rights

Named feature sections in setup teams and orgs.

Authored summary and link only; upstream rights retained.

[15]Nimbalyst · browser ↗1 / 41 claims
vendor-documentation · locator & rights

Named feature sections in browser.

Authored summary and link only; upstream rights retained.

[16]Nimbalyst · automations ↗1 / 41 claims
vendor-documentation · locator & rights

Named feature sections in automations.

Authored summary and link only; upstream rights retained.

[17]Nimbalyst · memory extension ↗1 / 41 claims
vendor-documentation · locator & rights

Key Features: hybrid retrieval and durable facts.

Authored summary and link only; upstream rights retained.