Developer Builds Logging System to Identify Unused Claude Code Subagents

·1 views

A developer discovered that 44 of their defined Claude Code subagents had gone unused over 30 days, prompting them to build a tracking system. The solution uses a Stop hook script that parses session transcripts at session end and appends invocation data to a JSONL log file. A scheduled daily report aggregates usage counts over 7-day and 30-day windows, producing a summary that highlights the top 10 agents and lists those with zero invocations. The system addresses a gap in Claude Code where there is no built-in mechanism to track whether defined agents are actually being called across sessions. By identifying these so-called zombie agents, developers can make evidence-based decisions about pruning their agent directories and reducing unnecessary context loaded at startup.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Sovereign Package Released Free on GitHub to Deploy Lemmings Across Cloud Regions

A developer has released Sovereign, a free open-source package on GitHub designed to deploy Lemmings load-testing tools across up to three cloud regions. The tool runs dry-runs to estimate costs and consolidates results into a single aggregated report as load test instances spin up and down. Sovereign is aimed at organizations using AI to build applications who want to stress-test their infrastructure before committing to large marketing or development budgets. The package is fully self-serve, requiring no involvement from the author, with guidance provided through detailed README files and code comments. The developer notes it is available now at no cost, with optional paid consulting support for those who need it.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer Builds DartShell to Fill WinSCP Gap for macOS Remote Workflows

A developer switching from Windows to macOS found no native WinSCP equivalent for remote file management, prompting them to build a tool called DartShell. WinSCP, widely used on Windows, supports SFTP, FTP, SCP, WebDAV, and S3 but has no official macOS version. DartShell aims to combine graphical SFTP and FTP file browsing with SSH access in a single interface, reducing the need to switch between terminal apps and separate file clients. The tool supports multiple connection profiles, remote directory browsing, and file transfers suited for regular server maintenance. The developer acknowledged that workflow preferences vary, with some users content using terminal commands like scp and rsync while others benefit from a GUI-based approach.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

How to Build a Feedback System That Stays Active Beyond Launch

A common problem with feedback channels is that they quickly become neglected once messages pile up without clear ownership or routing. A practical solution involves creating a small routing system that directs different request types — bugs, questions, feature ideas, security reports — to dedicated channels rather than a single inbox. Tools like GitHub Issue Forms, labeled workflows, a SUPPORT.md policy file, and a scheduled GitHub Action for overdue triage can keep the system functional over time. The approach works across open-source repositories, internal tools, small SaaS products, and community projects. The key principle is that every feedback route must have an explicit owner and a defined review schedule to remain sustainable.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer shares key design boundaries for building a responsible AI tarot tool

A developer building TaroMind AI, a free guided tarot experience, has outlined product design principles aimed at making AI-powered reflective tools safer and more predictable. The approach separates fixed, data-driven outputs from AI-generated explanations, ensuring the model clarifies rather than overrides structured answers. The design also prioritizes user privacy by keeping personal questions out of URLs, analytics payloads, and public logs, using tab-scoped session state instead. Clear disclaimers are recommended around sensitive topics such as medical, legal, financial, and mental-health decisions. The developer argues these boundaries apply broadly to AI products, not just tarot: critical decisions should live in typed data and code, while the model operates only within explicitly defined limits.

0 comments Read more at DEV Community