Plugins published to the Claude Code registry.

Automate every repetitive Android dev task — keystore creation, signing config, version bumping, Gradle version catalog management, new app scaffolding, and release checklists.

Built when the Android release checklist hit twelve manual steps at fleet scale. Encodes the entire Android release preparation layer — signing config generation, version management, Gradle catalog updates, pre-submission checklist — into repeatable commands. Manages 22+ Android apps in production under SudarshanTechLabs.

Kotlin code intelligence for Claude Code — automatic type checking, import resolution, null safety detection, and error diagnostics injected into Claude's context after every .kt file edit.

Bridges kotlin-language-server into Claude Code's context. After every .kt file edit, real diagnostics — type errors, missing imports, null safety violations — are injected automatically. Claude sees what the compiler sees and fixes issues in the same turn, before they reach the build.

Automate every Google Play Store publishing task — store listings, ASO, release notes, policy declarations, and screenshot copy — all from Claude Code.

Encodes the full Play Store submission workflow into structured commands. Every listing goes through the same character-limit format, ASO scoring pass, and compliance declarations checklist before every release. Built after Play Store compliance had no repeatable structure across 22+ apps.

Scan your AndroidManifest.xml and Gradle files, generate a Play Store-compliant privacy policy, and deploy it to GitHub Pages — all from Claude Code.

Automates the entire policy lifecycle: reads what your app actually collects (from the manifest and Gradle files), generates a compliant HTML policy, and deploys it to a predictable GitHub Pages URL. Built after manually maintaining policies across 22+ apps produced drift and missing URLs in Play Console.

Persistent memory for AI coding agents — extracts observations from each Claude Code session, injects relevant context back at the next session start. Pure Python, SQLite+FTS5, zero deps.

After every Claude Code session ends, extracts structured observations — bugs fixed, decisions made, errors seen — and stores them locally. At the next session start, injects relevant context as a Claude system message so the agent arrives knowing what broke last time, what decisions were made, and what's still pending. Available on PyPI as repomem.