PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteA Linux utility called worklog-for-claude-code turns Claude Code’s local session transcripts into a daily work log: activity grouped by project, token totals by day, account, and model, and a 14-day chart. Its creator, Adil Sadqi, describes it as a way to answer two practical questions at the end of a day: what work happened, and how much Claude Code used. It is a project description, not an independent security audit or software test.
What the work log records
The tool is designed for Claude Code sessions written to local transcripts. Sadqi says the generated report organizes work by project and shows token totals across days, accounts, and models. The report includes a 14-day chart. These are token counts, not dollar amounts: the project does not convert them into spend, and Pro and Max usage is subject to plan limits.
![Claude AI for Beginners Bible: [5 in 1] The Ultimate Guide to Automate Your Work, Save Hours Every Week, and Use AI for Real-World Results](https://m.media-amazon.com/images/I/415+fSJacsL._SL160_.jpg)
As an Amazon Associate I earn from qualifying purchases.
The recorded metadata is intentionally narrower than a transcript. Sadqi says the tool retains the opening prompt, truncated to 300 characters, along with token counts, file names, and the user’s own commit subjects. He says it does not retain file contents, Claude’s replies, tool output, or credentials.
How collection and reporting work
According to the project description, Claude Code’s local transcripts are read through session-start and session-end hooks. An hourly systemd timer is also intended to catch events a hook might miss. The tool writes a JSON record for each session to a Git repository controlled by the user. Each device writes to its own file paths, which the author says allows multiple machines to push into one repository without merge conflicts.
#1 Best Overall
A nightly job builds a Markdown report, which the local web interface displays. For use across computers or Claude accounts, the author suggests storing the data folder in a private GitHub or GitLab repository. That makes Git part of the user’s data-handling setup; the report does not imply that the project itself hosts or synchronizes the records.
Privacy design and its limits
Sadqi says the program removes patterns for API keys, tokens, JWTs, private keys, passwords in URLs, email addresses, and IP addresses before writing records. He also says there is no server or telemetry, and that data remains on the user’s machine and private repository. These are the creator’s stated design claims, not independently verified guarantees; redaction can only be relied on to the extent that its patterns recognize sensitive data.
AI summaries are optional and, according to the author, run through the user’s own claude -p. He says leaving that option off keeps the workflow fully local. The local interface is described as listening on loopback and checking Host and Origin headers, using CSRF tokens, and sending a strict Content-Security-Policy. Those implementation details are likewise author-reported, not the result of an external security review.
Recommended Free Tools
Trying it and setting it up
Preview the demo
The project article gives this command for launching a demo without first installing the package:
pipx run --spec worklog-for-claude-code claude-worklog demoThe demo interface is described as available at http://localhost:8766.
Install the service
The documented setup uses pipx, a Git repository created by the user, and the install-service command. The setup form is served at http://worklog.localhost:8765.
Rank #4
pipx install worklog-for-claude-code
claude-worklog install-serviceThe project article does not give a complete account of every environment-specific setup choice, so follow the current installation instructions in the author’s project article before relying on the service. For multiple machines, the author recommends using a private Git host for the shared data repository.
Who it fits—and what it does not cover
- Platform: The author lists Linux with systemd as the current requirement. A macOS version would need launchd service units.
- Assistant scope: It tracks Claude Code, not conversations on claude.ai, which are not written to the local transcripts the tool reads.
- Transcript compatibility: Claude Code’s transcript format is unofficial. Sadqi warns that an update could disrupt parsing until the parser is fixed; the described parser skips unrecognized lines rather than crashing.
- Usage interpretation: Tokens are not dollars. The project reports counts and does not establish a spend estimate or a plan-limit forecast.
The key trade-off is that the log depends on local Claude Code transcript data and Linux service infrastructure. It offers a private-repository workflow and useful daily summaries in the creator’s account, but its compatibility and privacy claims should be understood as project-reported rather than independently validated.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
