Skip to content
Sign in
  • Browse all toolsFilter by category and pricing
  • Browse categoriesWriting, code, image, audio and more
  • Compare toolsPut up to three side by side
  • Read the blogGuides and comparisons

agentsview

Local-first session search, analytics, and token cost tracking for AI coding agents.

agentsview is a local-first, open-source session search and analytics tool that archives coding-agent conversations and tracks token usage and compute costs. agentsview serves developers using Claude Code, Codex, and other coding agents, and is available under the MIT license through CLI installers, desktop downloads, Homebrew, or Docker.

Updated 9 Nov 20266 sections

agentsview website screenshot

What is agentsview?

agentsview is an open-source tool for browsing, searching, and analyzing AI coding-agent sessions. agentsview serves developers who want to revisit previous work and understand token consumption across Claude Code, Codex, and more than 50 other agents. The repository distributes agentsview under the MIT license.

agentsview uses a local-first archive: session history stays on the user's machine unless a feature that shares it is chosen. A local SQLite database supplies the web interface and usage reports, giving session browsing and cost analysis a shared source of archived data.

What agentsview can do

agentsview combines session discovery, search, usage reporting, and transcript-based inspection. Developers can find earlier work by label or linked pull request, narrow cost charts by project, and examine evidence of tool effectiveness. Reporting covers coding agents beyond Claude Code and includes filtering and machine-readable output.

  • agentsview searches archived coding-agent sessions.
  • agentsview tracks token consumption and compute costs.
  • agentsview calculates costs with prompt-cache creation and read tokens.
  • agentsview provides per-model breakdowns and agent filters.
  • agentsview exports usage reports as JSON.
  • agentsview exposes read-only MCP tools to connected assistants.
  • agentsview preserves retired-machine history through local archives.
  • agentsview syncs remote-machine sessions over SSH without installing a daemon remotely.

How agentsview works

agentsview discovers sessions from supported agents on first run, syncs them into a local SQLite database, and serves a web interface on loopback. Users can start the server in the foreground or run the same server as a background daemon. The web interface, API, session sync, and file watchers share a process.

agentsview offers command-line installers for macOS, Linux, and Windows, plus desktop downloads for macOS and Windows. Homebrew and Docker are additional installation paths. The desktop app and ordinary session CLI commands share a detached local daemon and start one when needed; daily usage reports use saved archive data.

Who agentsview is for

agentsview is for developers working with AI coding agents who need searchable histories and usage visibility across their tools. The documented workflows cover local development, archived machines, and remote environments. Command-line reporting also supports developers who want usage information in scripts, shell prompts, or status bars.

  • agentsview helps developers recover earlier work from session transcripts.
  • agentsview lets users inspect costs by agent, model, date, or project.
  • agentsview supports preserving history before retiring a machine.
  • agentsview supplies usage summaries for shell prompts and status bars.

agentsview pricing and access

agentsview is available as an MIT-licensed open-source project through its GitHub repository. The documented distribution options include installers, desktop releases, Homebrew, and a published Docker image. The source describes installation and licensing rather than subscription plans. Usage-cost reporting concerns coding-agent token consumption and compute costs, not an agentsview subscription.

agentsview strengths and limitations

agentsview combines a local session archive with cross-agent usage reporting and several deployment options. Automatic pricing uses LiteLLM and OpenRouter rates with an offline fallback, while reports account for prompt caching. Connected assistants can access the archive through read-only MCP tools rather than receiving write access through those tools.

  • agentsview in Docker requires agent session directories to be explicitly mounted and configured.
  • agentsview requires trusted-origin configuration for forwarded or remote browser access.
  • agentsview should have authentication enabled when exposed beyond loopback.
  • agentsview requires source syncing before saved reports include new source changes.

agentsview FAQ

What is agentsview?

agentsview is an open-source, local-first tool for searching coding-agent sessions and analyzing token usage and compute costs. agentsview supports Claude Code, Codex, and more than 50 other agents, using a local SQLite archive to power its web interface and reports.

Is agentsview open source?

agentsview is an open-source project distributed under the MIT license. agentsview is available from its GitHub repository, with documented installation options including command-line installers, desktop downloads for macOS and Windows, Homebrew, and a published Docker image.

Does agentsview keep coding sessions local?

agentsview keeps the session archive on the user's machine unless the user chooses a feature that shares it. agentsview discovers supported coding-agent sessions and syncs them into a local SQLite database used by the web interface and usage reports.

How does agentsview track token costs?

agentsview tracks token consumption and compute costs from its archived coding-agent sessions. agentsview uses LiteLLM and OpenRouter pricing rates with an offline fallback, accounts for prompt-cache creation and read tokens, and supports model breakdowns, agent filters, date filters, and JSON output.

Does agentsview support MCP and remote session sync?

agentsview provides eight read-only MCP tools for connected assistants to access the session archive. agentsview also supports syncing a remote machine over SSH without installing a daemon there and preserving a retired machine's history through local archives.

Can agentsview run in Docker?

agentsview offers a published Docker image and an included production Compose example. agentsview in a container can discover only explicitly mounted agent session directories with matching configuration. The image requires bearer-token authentication, and the documented examples publish the interface on loopback.

From the community

No reviews yet

Be the first to review this tool. What you write helps the next person choose.

More tools like agentsview

token-monitor

www.github.com

Local-first desktop tracking for AI coding token usage, costs, account limits, and multi-device activity.

Free

tokscale

www.github.com

Open-source CLI and dashboard for tracking AI coding agent token usage and costs.

Free

ccusage

www.github.com

Analyze coding agent token usage and costs from local data with unified CLI reports.

Free

CodeBurn

www.github.com

Track AI coding token usage, costs, quotas and potential savings from local session files.

Free