ronjunevaldoz

ronjunevaldoz/kmp-agent-skills

Plugin Claude Code

Kotlin Multiplatform agent skills, commands, and subagents for Claude Code.

ronjunevaldoz/kmp-agent-skills

Plugin Claude Code

AI agent skills for Kotlin Multiplatform (KMP) development. Covers clean architecture (6-layer model), feature scaffolding, Compose UI patterns, MVI, navigation, Ktor networking, SQLDelight, DataStore, dependency injection, testing, CI, and more. Use these skills to keep KMP architecture clean, repeatable, and easy to.

ronjunevaldoz/kmp-agent-skills

Hook Claude Code

Runs when a session starts, executing session-start-check-updates.sh via bash. From ronjunevaldoz/kmp-agent-skills.

ronjunevaldoz/kmp-agent-skills

Settings file Claude Code

Agent settings declaring 1 hook event (SessionStart).

ronjunevaldoz/kmp-agent-skills

Instructions file CodexOpenCode

AGENTS.md instructions for ronjunevaldoz/kmp-agent-skills, covering agent profile: staff jni bridge engineer & kmp architect, canonical sources — read before every task, skill routing, versioning & release — hard rules and operational guardrails.

ronjunevaldoz/kmp-agent-skills

Instructions file

Claude Code instructions for ronjunevaldoz/kmp-agent-skills, covering claude code cli project profile and automatically read the master agent instructions.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Runs architecture audits against a KMP project or the skills repo itself, interprets findings, and produces either a violation report or a prioritized adoption roadmap.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Generates consumer-facing changelogs and release notes from git history, per-skill ## Changelog sections, and commit metadata.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Shapes Kotlin Multiplatform and Compose UI work into a clear design direction before implementation begins. This agent is component-aware, accessibility-aware, and design-system-aware: it prefers existing KMP design skills first, then falls back to general design critique or UX…

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Keeps repo-facing documentation aligned with the actual repository shape, command set, and skill map.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Receives a specific list of blockers from the reviewer or validator and applies the minimum change to resolve each one — no refactoring, no new features, no architecture changes beyond the fix scope.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Reads accumulated lesson files from consumer projects or open GitHub issues, filters them, and proposes concrete amendments to source skills in this repo.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Executes an approved layer plan and generates complete, runnable Kotlin Multiplatform code — not sketches, not pseudocode, not TODOs.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline: a builder-first, stack-opinionated pipeline for Kotlin Multiplatform projects using Koin 4, Ktor 3, SQLDelight 2, AGP 9, and CMP 1.11.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Verifies a feature actually does what it was asked to do, by exercising it end to end — distinct from the validator (build/test gates pass) and the reviewer (architecture is correct). Both of those can pass on a feature that is wrong: tests only assert what someone thought to…

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Reviews implemented code against the 6-layer contract, Koin wiring rules, MVI contracts, and testTag coverage. The review is backed by auditproject.py — any finding from the script is an automatic blocker, not a warning.

ronjunevaldoz/kmp-agent-skills

Agent

Part of the KMP Agent Skills pipeline. Decides which skill to invoke and in what order for any KMP request. Routes to the right skill, builds the layer execution plan, and hands off to the implementer.

ronjunevaldoz/kmp-agent-skills

Command

KMP Agent Skills — check every screen for adaptive layout coverage gaps and redundant title violations across Compact (phone), Medium (tablet), and Expanded (desktop).

ronjunevaldoz/kmp-agent-skills

Command

Runs a visual-only audit of your app's Roborazzi screenshots using Claude vision. Detects design inconsistencies that have no code-level signal: spacing rhythm, color contrast, typography hierarchy, icon sizing, and alignment drift across screens.

ronjunevaldoz/kmp-agent-skills

Command

KMP Agent Skills — analyze Roborazzi golden screenshots for design consistency. Uses Claude's vision capability to inspect committed PNG files against the design system rules: color tokens, spacing, typography, TopAppBar structure, dark/light mode parity, and accessibility contrast.

ronjunevaldoz/kmp-agent-skills

Command

KMP Agent Skills — check whether the local skills collection is behind origin/main on github.com/ronjunevaldoz/kmp-agent-skills.

ronjunevaldoz/kmp-agent-skills

Command

Refactors code documentation across all architectural levels — classes, functions, extension functions, and inline blocks — per kmp-code-quality's Comment & KDoc Conventions ("By architectural level" section). Fixes file-by-file with per-file confirmation, the same pattern as /fix-design.

ronjunevaldoz/kmp-agent-skills

Command

Comprehensive project health doctor — heals docs sitemaps, git hooks, script permissions, and .agents/skills.lock.

ronjunevaldoz/kmp-agent-skills

Command

KMP Agent Skills — take a ticket from GitHub Issues (or any tracker) and ship a complete KMP feature implementation: branched, layered, Koin-wired, tested, and committed.