OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Evaluation & Observability

Agentacct

Agentacct is a local-first dashboard and command-line tool that organizes coding-agent session records into task histories, showing work steps, tool use, changed files, checks, elapsed time, tokens and estimated costs.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Agentacct — OpenViST project profile in Agent Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Agentacct when you want a local record of what coding agents did and what their recorded token use and estimated costs were, rather than relying only on an agent’s completion summary.

Who it is for

  • Developers who use coding agents and want to review their recorded activity across projects.
  • Teams or individual developers comparing agent work, checks and usage around project folders.

Documented uses

  • Trace a task’s recorded steps, tool calls, changed files and checks.
  • Compare work across agent sessions or agents within a project folder.
  • Review token usage and estimated costs by agent, model, provider or date.
  • Distinguish an agent-reported completion from a result supported by current passing checks.
Before you choose

Requirements & limits

The documented CLI requires Python 3.11 or later on macOS or Linux; Windows is supported through WSL. A native macOS app is available for macOS 14 or later. The README says no account is needed and data stays on the user’s computer.

  • Agentacct builds its views from local session logs; hooks and MCP servers bind when a new agent session starts after onboarding.
  • A reported completion is separate from a verified result, which requires current passing checks after the last recorded change.
  • Cost figures are pricing-table estimates, not invoices, and usage records may show confidence labels or gaps where joins are uncertain.
  • Session totals are shown by activity date; a session spanning multiple days is not split into exact daily consumption.
  • The project describes itself as early alpha.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Agentacct and start recording agent sessions

Install the CLI, initialize its local store and recorder, then begin a new agent session for hooks and MCP servers to bind.

Before you start

  • Python 3.11 or later on macOS or Linux, or Windows through WSL
  • pipx
  • A supported coding agent

Steps

  1. Install Agentacct with pipx install agentacct.

  2. Run agentacct onboard once on the machine to find agents, set up a local store and start the recorder.

  3. Open a new agent session so its hooks and MCP servers bind at session start.

  4. Run agentacct tui to open the terminal interface.

Expected outcome

A local Agentacct store and recorder are initialized, and the terminal interface can present recorded activity.

Guide limitations

  • Windows is supported only through WSL.
  • The project is described as early alpha.
  • Recorded costs are estimates rather than invoices.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
mikehasa/agentacct · Official homepage title · retained in its source language
How we review sourcesSuggest a correction