Automatic Time Tracking for Developers

    The time tracker that never asks you to start a timer.

    DevClocked captures every coding session and every AI agent run automatically, across your IDE, terminal, and browser. No timers. No screenshots. Just your real day, to the minute, with timesheets and invoices that write themselves.

    VS CodeCursorClaude CodeCodexTerminalChromeGitHubLinearJira

    Your real day, to the minute

    Every minute, counted automatically.

    94 minutes coding, three hours fighting config, four agents across three repos, the proof block no timer, Git graph, or token bill can give you. Captured at the source, reconciled while you sleep.

    Your day, reconciled

    Captured at the source. No timer, no Git graph, no token bill needed.

    Real coding

    94m

    Config & plumbing

    3h 04m

    Review & debug

    52m

    Agent runs

    4

    Repos touched

    3

    Ready to timesheet

    6h 41m

    Zero clicks to produce this. DevClocked watched the work and did the math.

    Why every other tracker falls short

    Manual timers were built for a world without AI agents.

    You don't fill in a timesheet at the end of a day pair-programming with four agents across three repos, you guess. DevClocked removes the guess.

    No timers to forget

    The tracking is automatic. Sessions across VS Code, Cursor, Terminal, and Chrome are detected and logged without a single click.

    AI work finally counts

    Agent runs, token spend, and the hours behind them show up as first-class work, the thing every other tracker is blind to.

    The toil becomes visible

    Config, debugging, repo-switching, and orchestration are separated from feature coding, so you see where the day actually went.

    Output, not surveillance

    No screenshots. No keystroke logging. We measure what shipped, not whether you were watched.

    Automatic capture

    Sessions that build themselves.

    The moment you start working, DevClocked opens a session and splits it into Work Blocks, each one mapped to the repo and project it belongs to, with every agent turn and its token cost attached. No timers, no tagging, no end-of-day reconstruction.

    Sessions that build themselves.

    Session analysis

    See how the work actually went.

    Every session is scored: a productivity signal, how aligned the work was to the project, and real code time versus config and context-switching. The detail a commit graph and a token bill can never show you, side by side.

    See how the work actually went.

    Outcome metrics

    Hours that land on the right ticket.

    Time is attributed automatically to the tickets, issues, and effort it advanced, so the work maps to outcomes, not just activity. What shipped, what it took, and where it counted, ready to bill or report.

    Hours that land on the right ticket.

    Set it and forget it

    Three steps. Then never think about it again.

    1

    Connect your surfaces

    Install the extension, the Mac app, or run npx devclocked setup for the terminal. Connect GitHub.

    2

    Code normally

    Write code, run agents, switch repos. The daemon watches; you don’t.

    3

    See your day reconciled

    Time Slice, Work Blocks, agent economics, and a ready-to-send timesheet appear automatically.

    Works where you already work

    Track every surface. Change nothing.

    One install per surface. One unified day across all of them.

    Trackers

    VS Code

    Capture edits, debugging, and assisted coding activity straight from VS Code as you work.

    Trackers

    Cursor Plugin

    Track Cursor sessions and agent-heavy edits without changing how developers code.

    Runtime

    Terminal / CLI

    Wrap terminal workflows, Codex, Claude Code, Aider, scripts, and daemon capture.

    Research

    Chrome Extension

    Tie docs, research, and browser context back to active engineering work.

    Output

    Codex Plugin

    Connect repositories, commits, and pull requests to the work blocks that produced them.

    Issues

    Claude Plugin

    Install the DevClocked app to sync issues and link shipped work to project evidence.

    Runtime

    MCP

    Query your live DevClocked signal: sessions, focus, agent runs, token spend, project context, and shipped output.

    npx devclocked setup

    Go deeper on any piece

    Every feature, explored in full.

    FAQ

    Frequently asked questions.

    The short answers for teams comparing Git analytics, token counters, and agent observability.

    Stop guessing where your day went.

    Install once. DevClocked counts every session and every agent run, automatically.

    Everything Git cannot show you.

    One operating layer for humans and agents shipping in parallel.