Agent bundles / developer
Code writing, git workflow, testing, and language/quality experts.
Understanding-oriented
Why the bundle is built this way.
developer is the largest bundle. It holds the git workflow (commit, pull request, worktree and Worktrunk skills), browser and devtools skills, game development, shells and UI development, and the reviewer agents.
Worktrunk owns every git worktree. Skills call wt, never git worktree, so Worktrunk hooks run on each create, merge and remove.
The worktree skill coordinates ticket work across Worktrunk worktrees, fresh agent sessions, review gates and durable records. Read how the worktree workflow works for its four diagrams.
Learning-oriented
Use the developer bundle for the first time.
/developer:commit in Claude Code, or /skill:commit in OMP.The agent cannot invoke this skill by itself. Only you can start it.
Task-oriented
Steps for one task each.
Skill: backlog-grooming
The backlog-grooming skill researches each ticket in a separate subagent, publishes one HTML report, and then asks which ticket changes to make.
/developer:backlog-grooming ABC-12 ABC-15 in Claude Code, or /skill:backlog-grooming ABC-12 ABC-15 in OMP.[ABC-12]: groomed 3 means the ticket is ready, with its story points.[ABC-15]: questions - 2 means two questions block sizing.[ABC-18]: cancel - Fixed in a1b2c3d means the ticket no longer applies.[ABC-21]: split - Two separate changes means another step must come first.The subagents never change code, tickets or branches. The skill reads and writes tickets through the reading-and-writing-tickets skill, so it works with any tracker that skill supports.
Skill: cleanup-merged
At session start, the cleanup-merged hook checks the repository at most once in 24 hours. When it finds branches or worktrees whose work is already in the default branch, the agent asks you in its first reply.
/developer:cleanup-merged in Claude Code, or /skill:cleanup-merged in OMP.The skill keeps unmerged, dirty and new work, and worktrees where a process still runs. It never deletes remote branches.
| Harness | Steps |
|---|---|
| Claude Code | Raise version in .claude-plugin/plugin.json and in the marketplace.json entry, then run: claude plugin update developer@zenobius-agent-bundles. Start a new session. |
| OMP | Start a new session. The symlink shows the change at once. |
| Pi | Start a new session. The symlink shows the change at once. |
| Zot | Start a new session. The symlink shows the change at once. |
Information-oriented
Read from the bundle files at build time.
| Part | Claude Code | OMP | Pi | Zot | Codex |
|---|---|---|---|---|---|
| Skillsskills/**/SKILL.md. Instructions that the agent or the user invokes. | supported | supported | supported | supported | not supported |
| Agentsagents/*.md. Subagent definitions with a role and a tool list. | supported | supported | supported | not supported | not supported |
| Hookshooks/hooks.json. Commands that run on harness events, such as SessionStart. | supported | supported | not supported | supported | not supported |
| Harness | Install | Reads |
|---|---|---|
| Claude Code | mise run claude.bootstrap (runs claude plugin install developer@zenobius-agent-bundles) | .claude-plugin/plugin.json, skills/, agents/*.md, hooks/hooks.json |
| OMP | mise symlinks ~/.omp/agent/bundles/developer to the bundle | package.json (pi key), skills/, agents/*.md, extensions/ |
| Pi | mise symlinks ~/.pi/agent/bundles/developer to the bundle | package.json (pi key), skills/, agents/*.md, extensions/ |
| Zot | mise symlinks ~/.local/state/zot/extensions/developer to the bundle | extension.json, skills/ |
146 skills
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| backlog-grooming | user and agent | yes | Grooms one or many backlog tickets. Fans out one research subagent per ticket to check each claim against the codebase, tests, LSP traces or E2E tests, then publishes a collated HTML report to shared context and asks which proposed ticket actions to apply. Use when the user says groom, refine, size, triage or qualify tickets, a sprint, an epic or a backlog query. |
| software-design | user and agent | — | Organizes reusable software design pattern references, when choosing a structural, creational, or behavioral pattern, resulting in faster selection of the right pattern-specific reference. |
| task-orchestrator | user and agent | — | Use when the user has established tasks and dependencies and needs parallel or serial delegation with separate implementation reviews. Schedules the graph and produces isolated worktrees, handoff files, implementation results, review results, and a final graph status. |
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| twg | user and agent | — | Root TWG CLI skill for Atlassian work-data tasks. Use typed commands for known anchors; use live `twg help` only when command shape or output contract is uncertain. |
| twg-agentic-search | user and agent | — | Use with root `twg` for deep iterative enterprise/company knowledge search and internal research with Rovo Search across connected apps/connectors including Confluence, Jira, Drive, Slack, Bitbucket, and GitHub. |
| twg-bench-lite | user and agent | yes | Run a lightweight single-prompt A/B comparison of free Atlassian/local MCP context vs TWG CLI graph context using paired read-only agent sessions. |
| twg-confluence | user and agent | — | Use with root `twg` for Confluence content, spaces, hierarchy, authoring, editing, comments, versions, permissions, exports, and CQL. Applies Confluence semantics and safe write rules. |
| twg-context-discovery | user and agent | — | Use with root `twg` for deep context, dependency maps, related entities, project-to-repo discovery, OOO catch-ups, and "catch me up" requests around a concrete anchor. |
| twg-engineering-work | user and agent | — | Use with root `twg` for code search, repositories using an API/package, implementation and reverse-dependency discovery, PR status and reviews, repo contributors, hot areas, and issue-to-PR lookups. |
| twg-jira | user and agent | — | Use with root `twg` for Jira workitems, projects, boards, sprints, fields, transitions, comments, links, and administration. Applies Jira semantics and safe mutation rules. |
| twg-jira-resolve-merged-work | user and agent | — | Clean up stale or unresolved Jira workitems. Dry-run single items, lists, boards, sprints, epics, or projects by matching Jira keys/titles to merged PRs, repos, Rovo/search-code hits, and assignee activity. |
| twg-operational-health | user and agent | — | Use with the root `twg` skill for on-call handoffs, incident response and investigation, post-incident reviews, reliability reviews, Assets refresh, capacity views, meeting summaries, and operational risk readouts. |
| twg-responsibility-routing | user and agent | — | Use with root `twg` to route owners, subject-matter experts, maintainers, reviewers, approvers, decision authorities, Heads of Engineering, or escalation paths for a topic, area, ask, project, service, or component. |
| twg-status-rollups | user and agent | — | Use with root `twg` for status rollups, personal work summaries, and decision-readiness or go/no-go briefs. Routes to `pr-tree`, `org-tree`, `work-tree`, or `workitem-tree`. |
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| a11y-debugging | user and agent | — | Uses Chrome DevTools MCP for accessibility (a11y) debugging and auditing based on web.dev guidelines. Use when testing semantic HTML, ARIA labels, focus states, keyboard navigation, tap targets, and color contrast. |
| agent-browser | user and agent | yes | Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. |
| browser-acceptance-evidence | user and agent | yes | Proves a change in a running app in five stages - reads the branch, ticket and work receipts, writes manual test steps as JSON and Markdown, confirms the plan with the user, drives it in a real browser with surf or Playwright for screenshots and optional video, then publishes the evidence to shared context and posts the steps to the issue with a link to the evidence. Use when a change needs manual test steps, QA steps, acceptance criteria, a smoke test, visual verification, "show me it works", or a hand-off that needs screenshots. Results in a confirmed plan, re-runnable scripts, a published evidence page and an issue comment instead of a claim in chat. |
| chrome-devtools | user and agent | — | Uses Chrome DevTools via MCP for efficient debugging, troubleshooting and browser automation. Use when debugging web pages, automating browser interactions, analyzing performance, or inspecting network requests. This skill does not apply to `--slim` mode (MCP configuration). |
| chrome-devtools-cli | user and agent | — | Use this skill to write shell scripts or run shell commands to automate tasks in the browser or otherwise use Chrome DevTools via CLI. |
| cookie-debugging | user and agent | — | Uses Chrome DevTools MCP for inspecting, debugging, and testing cookies, session state, authentication issues, and cookie consent compliance. Use when diagnosing 401/403 errors, authentication redirects, session expiration, Cookie/Set-Cookie header issues, cookie banner consent conformance, or third-party cookie/SameSite/Partitioned cookie warnings. |
| debug-optimize-lcp | user and agent | — | Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions "largest contentful paint", "page load speed", "CWV", or wants to improve how fast their hero image or main content renders. |
| firefox-devtools | user and agent | yes | Enables Firefox remote debugging workflows, when browser automation or protocol-level Firefox inspection is needed, resulting in configured Firefox RDP access and repeatable debugging steps. |
| memory-leak-debugging | user and agent | — | Diagnoses and resolves memory leaks in JavaScript/Node.js applications. Use when a user reports high memory usage, OOM errors, or wants to capture, compare, or inspect heap snapshots with Chrome DevTools MCP memory tools. |
| surf | user and agent | — | Controls Chrome through surf-cli for browser automation, debugging, and AI-assisted browsing. Use when tasks need navigation, form filling, screenshots, page inspection, network diagnostics, session-scoped multi-agent browsing, or multi-step browser workflows from the terminal. Results in deterministic, scriptable browser operations with less orchestration overhead. |
| troubleshooting | user and agent | — | Uses Chrome DevTools MCP and documentation to troubleshoot connection and target issues. Trigger this skill when list_pages, new_page, or navigate_page fail, or when the server initialization fails. |
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| bats-testing | user and agent | — | Use when writing shell-native tests for CLI tools, sourced Bash libraries, or REST APIs, especially when process boundaries, shell state, and exit/output behavior must be verified end-to-end. |
| code-comments | user and agent | — | Write clear, plain-spoken code comments and documentation that lives alongside the code. Use when writing or reviewing code that needs inline documentation—file headers, function docs, architectural decisions, or explanatory comments. Optimized for both human readers and AI coding assistants who benefit from co-located context. |
| codemapper | user and agent | — | Use when analyzing codebases for structure, finding symbols, tracing call paths, checking test coverage, or analyzing dependencies - provides instant AST-based code analysis using tree-sitter for Python, JavaScript, TypeScript, Rust, Java, Go, and C |
| create-new-bun-package-repo | user and agent | — | Use when creating new Bun packages from zenobi-us/bun-module template - automates repo creation, cloning, and setup using GitHub CLI; note setup.sh runs non-interactively with defaults requiring manual package.json updates |
| creating-cli-tools | user and agent | — | Design command-line interface parameters and UX: arguments, flags, subcommands, help text, output formats, error messages, exit codes, prompts, config/env precedence, and safe/dry-run behavior. Use when you’re designing a CLI spec (before implementation) or refactoring an existing CLI’s surface area for consistency, composability, and discoverability. |
| dead-code-sweep | user and agent | — | This skill should be used when cleaning up codebases that have accumulated dead code, redundant implementations, and orphaned artifacts — especially codebases maintained by coding agents. Triggers on "find dead code", "clean up unused code", "remove redundant code", "prune this codebase", "dead code sweep", "code cleanup", or when a codebase has gone through multiple agent-driven refactors and likely contains overlooked remnants. Systematically identifies cruft, categorizes findings, and removes confirmed dead code with user approval. |
| outline-wiki-reports | user and agent | — | Creates, structures, and updates evidence reports in Outline Wiki through its API. Use for QA results, acceptance evidence, audits, migrations, bug reproductions, and report hand-offs. Produces a verified document with linked checks, durable attachments, explicit coverage gaps, and revision-guarded updates. |
| publishing-with-release-please-and-moonrepo-on-github-actions | user and agent | — | Coordinates Release Please policy, Moon project selection, source-derived publish metadata, and SHA-pinned GitHub Actions publication. Use when standardizing monorepo release.yml/publish.yml workflows across main and release/* branches without passing computed versions or GitHub release tags between workflows. |
| release-please | user and agent | — | Use when adopting or operating Release Please for automated versioning, changelog generation, and GitHub releases in single-package or manifest-mode repositories |
| termscope | user and agent | — | Drive and inspect terminal applications programmatically. Use for TUI testing, visual regression, debugging rendering issues, and automating interactive CLI tools. Provides snapshot capture, keyboard/text input, text search, and a JSON-lines session protocol. |
| writing-reports | user and agent | yes | Writes a self-contained HTML evidence report with screenshots, captions, result badges and a clickable summary table. Use when work must be proved rather than claimed - manual test results, acceptance evidence, an audit, a migration check, a bug reproduction, or any hand-off that needs "here is the screenshot that shows it". Results in one portable directory that renders anywhere, with gaps and leftover test data named. |
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| 2d-essentials | user and agent | — | Use when working with 2D-specific systems — TileMaps, parallax scrolling, 2D lights and shadows, canvas layers, particles 2D, custom drawing, and 2D meshes in Godot 4.3+ |
| 3d-essentials | user and agent | — | Use when working with 3D-specific systems — materials, lighting, shadows, environment, global illumination, fog, LOD, occlusion culling, and decals in Godot 4.3+ |
| ability-system | user and agent | — | Use when building character abilities — Resource-based abilities with cost/cooldown/cast, buffs/debuffs, stat modifiers, gameplay tags, and HUD binding |
| addon-development | user and agent | — | Use when creating Godot editor plugins — EditorPlugin, @tool scripts, custom inspectors, and dock panels |
| ai-navigation | user and agent | — | Use when implementing AI movement — NavigationAgent2D/3D, steering behaviors, behavior trees, and patrol patterns |
| animation-system | user and agent | — | Use when implementing animations — AnimationPlayer, AnimationTree, blend trees, state machines, sprite animation, and code-driven animation |
| assets-pipeline | user and agent | — | Use when importing and managing assets — image compression, 3D scene import, audio formats, resource formats, and import configuration |
| audio-system | user and agent | — | Use when implementing audio — audio buses, AudioStreamPlayer, spatial audio, music management, SFX pooling, and dynamic mixing |
| beehave | user and agent | — | Use when using the Beehave addon — pure-GDScript behavior trees with composites, decorators, leaves, a blackboard, and a visual runtime debugger |
| camera-system | user and agent | — | Use when implementing cameras — smooth follow, screen shake, camera zones, and transitions for 2D and 3D |
| component-system | user and agent | — | Use when building reusable node components — composition patterns, component communication, and interface design |
| csharp-godot | user and agent | — | Use when working with C# in Godot — conventions, GodotSharp API differences from GDScript, project setup, and interop |
| csharp-signals | user and agent | — | Use when implementing signals in C# — [Signal] delegates, EmitSignal patterns, async signal awaiting, and event-driven architecture |
| cyberpunk-help | user and agent | — | Help with a Cyberpunk 2077 playthrough on Windows, Linux, or macOS (Vortex or manual mods): missions, Night City locations, save state, and CET console commands. On first use in a conversation (or when the deploy list changes / the user says they added mods), discover the local install by signature files, read OS + today's date, then check Nexus for newer files and relevant comments. Never freeze a mod list, save path, or CET cheat-sheet in this skill. Use when the user asks about CP2077 quests, map places, "dónde está", Afterlife, Nocturne, journal, CET, Cyber Engine Tweaks, or runs /cyberpunk-help. |
| cyberwise | user and agent | — | Front door for diagnosing a modded Cyberpunk 2077 install - the method rules that apply to every CP2077 modding task, where each kind of mod lives on disk, and which cyberwise-* skill covers the problem at hand. Use at the start of any Cyberpunk 2077 modding question, and when unsure which part applies. |
| cyberwise-backstory | user and agent | — | Help someone build a Cyberpunk 2077 character - V's backstory, voice, and the decisions they would make - by interviewing them rather than writing it for them. Knows the lifepath prologues, Night City geography and the game's own document formats well enough to ask good questions. Use when someone wants help with their V's history, personality, name, look, roleplay decisions, or a character dossier; and when they ask what is canon. |
| cyberwise-conflicts | user and agent | — | Cyberpunk 2077 load order and mod conflicts - why earlier in modlist.txt wins, detecting archives that are installed but contributing nothing, and visual bugs that are not conflicts at all. Use when a mod is installed and enabled but does nothing, when textures or body parts look wrong or mismatched, when two mods fight, or when reading .archive internals. |
| cyberwise-crashes | user and agent | — | Diagnose Cyberpunk 2077 crashes, hangs and failures to launch on a modded install - which log actually says what, how to bisect a large mod list without wasting launches, and why the obvious memory reading is usually wrong. Use when the game crashes, hangs, closes silently, or stops launching after a mod change. |
| cyberwise-feedback | user and agent | — | Help someone report a problem with Cyberwise itself, or correct something it got wrong - gather the facts the author needs, write it up, and send it to the right place (Discord or a GitHub issue). Use when someone says a cyberwise skill or tool is broken, gave wrong advice, or contradicts what they see in game; when they ask how to report a bug, request a feature, or contact the author; and when you yourself have just got something wrong and they are visibly annoyed. |
| cyberwise-hotkeys | user and agent | — | Find what every key is actually bound to on a modded Cyberpunk 2077 install, across the five separate stores that hold bindings, and generate a hotkey cheatsheet from what is really on disk. Also reads a Corsair iCUE mouse profile and joins each thumb button to the keystroke it sends and the binding that keystroke actually has. Use when asked what a key or a mouse button does, whether a key is free, how to rebind something, why a mouse button does nothing in game, or for a keybind reference or cheatsheet. |
| cyberwise-modbase | user and agent | — | Build and maintain a per-user knowledge base covering every mod actually deployed on this install - the inventory that comes from the deployment manifest rather than a manager's list, how to derive a Nexus id without guessing one, and the order to document them in. Use when asked what a mod does, when auditing a large load order, or when starting a documentation pass. |
| cyberwise-netsec | user and agent | — | Turn NetSec's diagnostic log into places you can stand - parse the GAP lines it writes for networks with no way in, cluster them into candidate access-point sites, and write them as CETMonkey teleport waypoints. Use when asked where NetSec found holes in the world, to plan where to place access points, or to generate teleport waypoints from a NetSec log. |
| cyberwise-recommends | user and agent | — | Tell someone a mod is missing only when it actually blocks what they are doing, and never twice - the capability checks behind "you cannot do that without X", and the on-disk preference that records a no. Use when a requested task needs a tool the install does not have, and before mentioning any mod nobody asked about. |
| cyberwise-reports | user and agent | — | Generate deliverables about a modded Cyberpunk 2077 install - a system profile that says what is likely wrong, a full mod-list inventory with descriptions, and the house style plus headless verification for any HTML or Discord-markdown report. Use when someone reports a problem with no detail, when asked what mods are installed, or when generating any page or report for a user to read. |
| cyberwise-reshade | user and agent | — | ReShade on Cyberpunk 2077 - why some mods need the add-on build rather than the standard one, and how shader packs collide when several are stacked. Use when ReShade effects fail to load, when shaders error on compile, or when combining multiple shader packs. |
| cyberwise-saves | user and agent | — | Read Cyberpunk 2077 save files and character appearance data - decompressing a save, locating the appearance section, and decoding AppearanceChangeUnlocker (ACU) presets. Use when asked what a character's appearance settings are, to recover appearance data without re-recording it in game, or to read anything out of a .dat save. |
| cyberwise-sitebuilder | user and agent | — | Turn a folder of character documents into a website - plain HTML, CSS and JS with no build step, no runtime and no account, deployable by copying a folder. Each character gets its own stylesheet, CSS Zen Garden style, so a dossier and a monologue do not come out looking like the same page. Use when someone wants to publish or share their V's backstory, character records or install documentation as a web page rather than a Markdown file. |
| cyberwise-tweaks | user and agent | — | Author Cyberpunk 2077 mods that are code rather than assets - TweakXL/TweakDB records, CET Lua, and redscript that changes world state. Covers finding the real record ID rather than guessing it, what the CET console can and cannot do, reading the game's own shipped script dump for real signatures, and locating in-game text. Use when writing or repairing a .yaml tweak or a .reds script, running console commands, or hunting for an in-game string or record. |
| cyberwise-wiki | user and agent | — | Store what you learn about this game and this install as an OKF knowledge bundle instead of burying it in a skill file - the article format, the two bundles and why one never ships, and the validator that enforces the boundary. Use before writing anything down that a later session would want to look up, and whenever a skill file is growing a section that is really reference material. |
| dedicated-server | user and agent | — | Use when building dedicated servers — headless export, server architecture, lobby management, and deployment |
| dependency-injection | user and agent | — | Use when managing dependencies between systems — autoloads, service locators, @export injection, and scene injection patterns |
| dialogue-manager | user and agent | — | Use when using the Dialogue Manager addon — .dialogue files with titles, responses, conditions and mutations, runtime balloons, and C# support |
| dialogue-system | user and agent | — | Use when implementing dialogue — data structures for branching dialogue, conditions, and UI presentation |
| event-bus | user and agent | — | Use when implementing decoupled communication between nodes — global EventBus autoload with typed signals |
| export-pipeline | user and agent | — | Use when exporting and distributing Godot games — export presets, platform settings, CI/CD with GitHub Actions |
| gdextension | user and agent | — | Use when building native extensions for Godot — godot-cpp (C++) or gdext (Rust), binding classes, building, and GDScript/C# interop |
| gdscript-advanced | user and agent | — | Use when writing production-grade GDScript — performance idioms, metaprogramming, @tool lifecycle, async pitfalls, signal/Callable trade-offs, profiler-driven idioms, and common pitfalls |
| gdscript-patterns | user and agent | — | Use when writing GDScript — static typing, await/coroutines, lambdas, match patterns, export annotations, inner classes, and common idioms |
| godot-brainstorming | user and agent | — | Use when designing a new Godot feature or system — guides scene tree planning, node type selection, and architectural decisions |
| godot-code-review | user and agent | — | Use when reviewing GDScript or C# Godot code — checklist of best practices, common anti-patterns, and Godot-specific pitfalls |
| godot-debugging | user and agent | — | Use when debugging Godot projects — remote debugger, print techniques, signal tracing, common error patterns and fixes |
| godot-mentor | user and agent | — | Use when the user wants to learn Godot while building — teaching mode that explains the concept, the editor setup, and what to verify, instead of just delivering code. Triggers on "teach me", "explain as we go", "I'm learning Godot", "guide me", "walk me through", "help me understand". |
| godot-optimization | user and agent | — | Use when optimizing Godot games — profiler, draw calls, physics tuning, memory management, and common bottlenecks |
| godot-project | user and agent | yes | Use when starting a session in a Godot project — detects the project, engine version, renderer, language, mentor mode, and agent-instructions wiring |
| godot-project-setup | user and agent | — | Use when creating a new Godot 4.x project — scaffolds recommended directory structure, project settings, autoloads, and .gitignore |
| godot-testing | user and agent | — | Use when writing tests for Godot projects — TDD workflow with GUT and gdUnit4, covers both GDScript and C# |
| godot-ui | user and agent | — | Use when building user interfaces — Control nodes, themes, anchors, containers, and layout patterns |
| hud-system | user and agent | — | Use when building in-game HUDs — health bars, score displays, minimap, notifications, and damage numbers |
| input-handling | user and agent | — | Use when implementing input — InputEvent system, Input Map actions, controllers/gamepads, mouse/touch, action rebinding, and input architecture |
| inventory-system | user and agent | — | Use when building inventory systems — Resource-based items, slot management, stacking, and UI binding |
| limboai | user and agent | — | Use when using the LimboAI addon — behavior trees and hierarchical state machines (C++ GDExtension) with a visual editor, BTTask subclassing, and a blackboard |
| localization | user and agent | — | Use when implementing localization (i18n/l10n) — TranslationServer, CSV/PO translation files, locale switching, RTL support, and pluralization in Godot 4.3+ |
| math-essentials | user and agent | — | Use when implementing game math — vectors, transforms, interpolation, curves, random number generation, and common geometric recipes |
| mobile-development | user and agent | — | Use when targeting Android/iOS — export and signing, permissions, plugins, in-app purchases, ads, app lifecycle, device features, and mobile performance |
| multiplayer-basics | user and agent | — | Use when implementing multiplayer — MultiplayerAPI, ENet/WebSocket peers, RPCs, and authority model |
| multiplayer-sync | user and agent | — | Use when synchronizing multiplayer state — MultiplayerSynchronizer, interpolation, prediction, and lag compensation |
| multithreading | user and agent | — | Use when running work off the main thread — WorkerThreadPool, Thread/Mutex/Semaphore, call_deferred, thread-safe scene access, and threaded resource loading |
| particles-vfx | user and agent | — | Use when implementing particle effects — GPUParticles2D/3D, ParticleProcessMaterial, emission shapes, subemitters, trails, attractors, collision, and common VFX recipes |
| phantom-camera | user and agent | — | Use when using the Phantom Camera addon — PhantomCamera2D/3D with priority-based switching, follow and look-at modes, and tween transitions |
| physics-system | user and agent | — | Use when working with physics bodies, collision shapes, raycasting, areas, rigid bodies, ragdolls, soft bodies, Jolt physics, and physics interpolation in Godot 4.3+ |
| player-controller | user and agent | — | Use when implementing player movement — CharacterBody2D/3D patterns, input handling, physics, common movement recipes |
| popochiu | user and agent | — | Use when using the Popochiu addon — point-and-click adventure framework with rooms, characters, props/hotspots, inventory, dialog trees, and a command-based GUI |
| procedural-generation | user and agent | — | Use when implementing procedural generation — noise-based terrain, BSP dungeons, cellular automata caves, wave function collapse, and seeded randomness in Godot 4.3+ |
| resource-pattern | user and agent | — | Use when creating data containers in Godot — custom Resources for configuration, items, stats, and editor integration |
| responsive-ui | user and agent | — | Use when handling multiple resolutions — stretch modes, aspect ratios, DPI scaling, and mobile/desktop adaptation |
| save-load | user and agent | — | Use when implementing save/load systems — ConfigFile, JSON, Resource serialization, save game architecture |
| scene-organization | user and agent | — | Use when designing scene tree structure — composition vs inheritance, when to split scenes, node hierarchy patterns |
| shader-basics | user and agent | — | Use when implementing shaders — Godot shader language, visual shaders, common visual recipes, and post-processing effects |
| shader-effects | user and agent | — | Visual shader effects—glow/bloom, chromatic aberration, distortion, vignette, film grain, scanlines, glitch, dissolve, outline, and fresnel. Use when adding visual polish, post-processing effects, or stylized rendering to shaders. |
| shader-fundamentals | user and agent | — | GLSL shader fundamentals—vertex and fragment shaders, uniforms, varyings, attributes, coordinate systems, built-in variables, and data types. Use when writing custom shaders, understanding the graphics pipeline, or debugging shader code. The foundational skill for all shader work. |
| shader-noise | user and agent | yes | Procedural noise functions in GLSL—Perlin, simplex, Worley/cellular, value noise, FBM (Fractal Brownian Motion), turbulence, and domain warping. Use when creating organic textures, terrain, clouds, water, fire, or any natural-looking procedural patterns. |
| shader-router | user and agent | — | Decision framework for GLSL shader projects. Routes to specialized shader skills (fundamentals, noise, SDF, effects) based on task requirements. Use when starting a shader project or needing guidance on which shader techniques to combine. |
| state-machine | user and agent | — | Use when implementing state machines in Godot — enum-based, node-based, and resource-based FSM patterns with trade-offs |
| ttrpg-gm | user and agent | — | TTRPG Game Master for dark, mature campaigns. Use when the user wants to play, run, start, continue, or resume a tabletop RPG, campaign, or adventure. Supports Cyberpunk 2077, Witcher, Warhammer 40K, Vampire the Masquerade, Shadowrun, Altered Carbon, Blade Runner, Mass Effect, Fallout, Dune, Game of Thrones, Star Wars, and custom universes. Features: autonomous NPCs with hidden secrets, dual-consequence system tracking world state AND relationships, hidden D20 rolls with graduated feedback, psychological gauges, cinematic combat, campaign persistence across sessions. |
| tween-animation | user and agent | — | Use when implementing tweens — property animation, method tweening, chaining, parallel sequences, easing, and common UI/gameplay motion recipes |
| using-godot-prompter | user and agent | — | Bootstrap skill — establishes how to find and use GodotPrompter skills, with platform-specific tool mapping |
| webgl | user and agent | — | WebGL shaders and effects for JARVIS 3D HUD |
| xr-development | user and agent | — | Use when building VR/AR/XR applications — OpenXR setup, XROrigin3D, hand tracking, controllers, passthrough, and Meta Quest deployment in Godot 4.3+ |
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| cleanup-merged | user and agent | yes | Removes local branches and worktrees whose work is already in the default branch, including squash merges, when the user asks to clean up, prune, or tidy finished branches or worktrees, or when a cleanup-merged session-start notice lists candidates, resulting in a previewed, user-confirmed removal through Worktrunk that keeps busy, dirty, young, and unmerged work. |
| commit | user only | — | Commit staged and unstaged changes with atomic conventional commits. |
| pr | user and agent | yes | Runs the git and GitHub pull request workflow of any repository - opens and updates pull requests, names and creates branches, makes semantic commits, watches a PR for failed checks and reviewer comments, and resolves review threads with the gh CLI. Use when opening or updating a PR, writing a PR body, committing work, naming or creating a branch, watching a PR, working through review comments or failed checks, or posting a PR line comment. Results in conventional branches and commits, a reviewer-ready PR body, green checks and zero unresolved review threads. |
| worktree ↗ | user and agent | yes | Route worktree phases and dependency views to their playbooks. Use the goal workflow in a guided-goal interview to complete an initiative or explicit ticket collection. |
| worktrunk | user and agent | — | Guidance for Worktrunk (the `wt` CLI) — git worktree management, hooks, and config. Load when editing .config/wt.toml or ~/.config/worktrunk/config.toml; adding, modifying, or debugging hooks (post-merge, post-start, pre-commit, pre-merge, post-switch, etc.); configuring commit message generation or command aliases; or troubleshooting wt behavior. Also answers general worktrunk/wt questions. |
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| hrdx | user and agent | — | Reference for hrdx (https://github.com/patriceckhart/hrdx), a terminal multiplexer for AI coding agents (Codex, Claude Code, pi, zot). Use when adding or editing hrdx themes, remapping keys (keys.json), registering a custom agent CLI (harness.json), custom sounds (sounds.json), wiring hrdx config into this repo's mise symlinks, or answering hrdx CLI flag / socket API questions. Covers exact config file schemas, default keymap actions, and where files live in this dotfiles repo. |
| hrdx-plugins | user and agent | — | Use when creating, approving, debugging, or driving hrdx plugins and the hrdx socket API; produces safe plugin manifests, bounded JSON-line protocol handlers, least-privilege grants, and tested lifecycle or workspace operations. |
| hrdx-subagents | user and agent | — | Uses hrdx as the subagent framework for parallel agent work, independent implementation, focused review, and isolated worktrees, with Codex, Claude Code, Pi, Zot, or registered custom harnesses. |
| lazy-nvim-optimization | user and agent | — | Profiles Neovim startup performance and optimizes lazy.nvim plugin loading with targeted lazy-loading specs, priorities, and event triggers. Contains profiling workflow and bottleneck checklists. Make sure to use this skill whenever the user mentions slow neovim startup, neovim takes time to load, plugin profiling, lazy-loading, lazy.nvim config, startup time optimization, or neovim performance — even if the user doesn't say "lazy.nvim" explicitly. |
| neovim-best-practices | user and agent | — | Configures Neovim with the user's opinionated Lua patterns, namespaced config structure, LSP via Mason + nvim-lspconfig, and leader key conventions. Contains user-specific decisions and checklists Claude cannot know without this skill. Make sure to use this skill whenever the user mentions neovim, nvim, init.lua, LSP setup, keymaps, autocommands, neovim plugins, Mason, TypeScript LSP, slow LSP, neovim config, or any neovim question — even if Claude thinks it can answer from general knowledge. |
| zellij | user and agent | — | Use when manipulating Zellij sessions, creating tabs or panes, or looking up Zellij CLI commands for terminal multiplexer operations |
| Skill | Invoked by | Scripts | Description |
|---|---|---|---|
| basic-design-principles | user and agent | — | Enforce a precise, minimal design system inspired by Linear, Notion, and Stripe. Use this skill when building dashboards, admin interfaces, or any UI that needs Jony Ive-level precision - clean, modern, minimalist with taste. Every pixel matters. |
| cyberpunk-ui | user and agent | — | Web and App implementation guide for Cyberpunk UI. Trigger when user wants neon colors, dark backgrounds, high-tech dystopian aesthetics, and hacking interfaces. |
| feature-arch | user and agent | — | React feature-based architecture guidelines for scalable applications (formerly feature-architecture). This skill should be used when writing, reviewing, or refactoring React code to ensure proper feature organization. Triggers on tasks involving project structure, feature organization, module boundaries, cross-feature imports, data fetching patterns, or component composition. |
| framer-motion-animator | user and agent | — | Creates smooth animations and micro-interactions using Framer Motion including page transitions, gestures, scroll-based animations, and orchestrated sequences. Use when users request "add animation", "framer motion", "page transition", "animate component", or "micro-interactions". |
| frontend-design | user and agent | — | Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications. Generates creative, polished code that avoids generic AI aesthetics. |
| frontend-patterns | user and agent | — | Frontend development patterns for React, Next.js, state management, performance optimization, and UI best practices. Use when building or reviewing React or Next.js components, state, or render performance. |
| gsap-framer-scroll-animation | user and agent | — | Use this skill whenever the user wants to build scroll animations, scroll effects, parallax, scroll-triggered reveals, pinned sections, horizontal scroll, text animations, or any motion tied to scroll position — in vanilla JS, React, or Next.js. Covers GSAP ScrollTrigger (pinning, scrubbing, snapping, timelines, horizontal scroll, ScrollSmoother, matchMedia) and Framer Motion / Motion v12 (useScroll, useTransform, useSpring, whileInView, variants). Use this skill even if the user just says "animate on scroll", "fade in as I scroll", "make it scroll like Apple", "parallax effect", "sticky section", "scroll progress bar", or "entrance animation". Also triggers for Copilot prompt patterns for GSAP or Framer Motion code generation. Pairs with the premium-frontend-ui skill for creative philosophy and design-level polish. |
| playwright | user and agent | — | Playwright testing best practices for Next.js applications (formerly test-playwright). This skill should be used when writing, reviewing, or debugging E2E tests with Playwright. Triggers on tasks involving test selectors, flaky tests, authentication state, API mocking, hydration testing, parallel execution, CI configuration, or debugging test failures. |
| react | user and agent | — | React 19 performance optimization guidelines for concurrent rendering, Server Components, actions, hooks, and memoization (formerly react-19). This skill should be used when writing React 19 components, using concurrent features, or optimizing re-renders. This skill does NOT cover Next.js-specific features like App Router, next.config.js, or Next.js caching (use nextjs-16-app-router skill). For client-side form validation with React Hook Form, use react-hook-form skill. |
| react-doctor | user and agent | — | Use when finishing a feature, fixing a bug, before committing React code, or when the user types `/doctor`, asks to scan, triage, or clean up React diagnostics. Covers lint, accessibility, bundle size, architecture. Includes a regression check and a full local-triage workflow that fetches the canonical playbook. |
| react-hook-form | user and agent | — | React Hook Form performance optimization for client-side form validation using useForm, useWatch, useController, and useFieldArray. This skill should be used when building client-side controlled forms with React Hook Form library. This skill does NOT cover React 19 Server Actions, useActionState, or server-side form handling (use react-19 skill for those). |
| react-patterns | user and agent | — | React 18/19 patterns including hooks discipline, server/client component boundaries, Suspense + error boundaries, form actions, data fetching, state management decision trees, and accessibility-first composition. Use when writing or reviewing React components. |
| shadcn | user and agent | — | shadcn/ui component library best practices and patterns (formerly shadcn-ui). This skill should be used when writing, reviewing, or refactoring shadcn/ui components to ensure proper architecture, accessibility, and performance. Triggers on tasks involving Radix primitives, Tailwind styling, form validation with React Hook Form, data tables, theming, or component composition patterns. |
| shadcn-installing-components | user and agent | — | Installs shadcn registry components into ui-foundation with monorepo-safe paths, when adding official or third-party shadcn components to pkgs/libs/ui-foundation, resulting in registry-folder placement, relative imports, explicit subpath exports, and no root barrel exports. |
| tailwind | user and agent | — | Tailwind CSS v4 performance optimization and best practices guidelines (formerly tailwindcss-v4-style). This skill should be used when writing, reviewing, or refactoring Tailwind CSS v4 code to ensure optimal build performance, minimal CSS output, and correct usage of v4 features. Triggers on tasks involving Tailwind configuration, @theme directive, utility classes, responsive design, dark mode, container queries, or CSS generation optimization. |
| tanstack-query | user and agent | — | TanStack Query v5 performance optimization for data fetching, caching, mutations, and query patterns. This skill should be used when using useQuery, useMutation, queryClient, prefetch patterns, or TanStack Query caching. This skill does NOT cover generating query hooks from OpenAPI (use orval skill) or mocking API responses in tests (use test-msw skill). |
| test-driven-development | user and agent | — | Test-Driven Development methodology and red-green-refactor workflow (formerly test-tdd). This skill should be used when practicing TDD, writing tests first, designing tests before implementation, or reviewing test-first approaches. Triggers on "write tests first", "test before code", "red green refactor", "test driven development". This skill does NOT cover Vitest framework specifics (use vitest skill) or API mocking with MSW (use msw skill). |
| typescript | user and agent | — | This skill should be used when the user asks to "optimize TypeScript performance", "speed up tsc compilation", "configure tsconfig.json", "fix type errors", "improve async patterns", or encounters TS errors (TS2322, TS2339, "is not assignable to"). Also triggers on .ts, .tsx, .d.ts file work involving type definitions, module organization, or memory management. Does NOT cover TypeScript basics, framework-specific patterns, or testing. |
| ui-design | user and agent | — | UI/UX and frontend design best practices guidelines (formerly frontend-design). This skill should be used when writing, reviewing, or designing frontend code to ensure accessibility, performance, and usability. Triggers on tasks involving HTML structure, CSS styling, responsive layouts, form design, animations, or accessibility improvements. |
| ux-preview | user and agent | — | Use when an agent needs to preview UX options from a requested count and variation parameters before implementation; produces a locally served HTML comparison report with honest prototype flows. |
| vitest | user and agent | — | Vitest testing framework patterns for test setup, async testing, mocking with vi.*, snapshots, and test performance (formerly test-vitest). This skill should be used when writing or debugging Vitest tests. This skill does NOT cover TDD methodology (use test-tdd skill), API mocking with MSW (use test-msw skill), or Jest-specific APIs. |
| Agent | Tools | Description |
|---|---|---|
| godot-animator | all tools | Use this agent when the user needs to design or implement animation systems in Godot 4.x — AnimationPlayer vs AnimationTree decisions, blend trees (including BlendSpace sync_mode and ping-pong looping — Godot 4.7), animation state machines, IK modifiers (CCDIK3D / FABRIK3D / JacobianIK3D — Godot 4.6+), BoneConstraint3D, mocap retargeting, sprite vs skeletal animation, and procedural animation. The agent distinguishes animation FSM (inside AnimationTree) from gameplay FSM (state-machine skill), and hands gameplay logic back to game-dev. |
| godot-code-reviewer | all tools | Use this agent when the user wants their Godot GDScript or C# code reviewed for best practices, anti-patterns, performance issues, or Godot-specific pitfalls. Also use when completing a major feature and wanting a quality check. |
| godot-csharp-engineer | all tools | Use this agent for C#-first Godot 4.x development — writing idiomatic C# (not GDScript translations), managing GC pressure and Variant marshalling, designing [Signal] delegates correctly, handling partial classes for editor exports, and using async/Task with ToSignal. Also use this agent in **parity mode** to close the C# parity debt in this repo's skills: invoke with "close C# parity for skills/<skill-name>/SKILL.md" or similar phrasing, and the agent will read the parity debt notes, write the missing C# block per the per-section guidance, run the validator, and update the notes file. |
| godot-game-architect | all tools | Use this agent when the user needs help with Godot 4.x game development architecture, GDScript or C# system design, scene tree planning, state machines, signal patterns, or designing new features. This includes planning new features, designing game systems, refactoring existing code, debugging architectural issues, or creating implementation plans. |
| godot-game-dev | all tools | Use this agent when the user needs help implementing Godot Engine features, including GDScript or C# coding, scene/node setup, player controllers, enemy AI, inventory systems, dialogue, save/load, HUD, cameras, multiplayer, or any Godot-specific implementation. |
| godot-performance-profiler | all tools | Use this agent when the user reports lag, stutter, frame drops, draw call spikes, GC pauses (C#), physics slowdowns, or any other performance issue in their Godot 4.x project. The agent reads code, asks for profiler captures before guessing, classifies the bottleneck (CPU vs. GPU, draw calls vs. fillrate, physics vs. scripts, GC), and prescribes fixes grounded in the godot-optimization skill. |
| godot-shader-author | all tools | Use this agent when the user needs to write a custom Godot shader, post-processing effect, screen-space effect, custom material, visual shader graph, or Compositor effect. The agent is a specialist in the Godot shader language, knows when to reach for visual shaders vs. text shaders, and picks the right shader type for the task. |
| godot-tools-engineer | all tools | Use this agent for Godot 4.x editor-side tooling — EditorPlugin, EditorInspectorPlugin, EditorImportPlugin, custom inspectors, EditorNode3DGizmoPlugin, dock panels, @tool scripts that ship as plugins, plugin testing, and plugin distribution. Works in both GDScript and C# (with #if TOOLS guards). GDExtension (C++ native modules) is explicitly out of scope and handled by a different agent. |
| godot-ui-designer | all tools | Use this agent when the user needs to build or refactor user interfaces in Godot 4.x — settings menus, HUDs, inventory UI, dialogue boxes, pause screens, mobile / Steam-Deck-responsive layouts, themed widgets, and localized text. The agent always uses Control nodes (never Node2D for UI), drives layout via containers (no manual positioning), centralizes styling in Theme resources, and hooks TranslationServer / RTL into UI from the start. Knows Godot 4.5+ additions: FoldableContainer, Stacked Label Effects, custom_maximum_size and VirtualJoystick (4.7). |
| reviewer | read, grep, find, ls, bash | Code review specialist that validates implementation and fixes issues |
| reviewer-arch | all tools | Architecture reviewer for design reviews, module boundary validation, abstraction evaluation, and technical debt assessment. Does NOT write code. |
| reviewer-correctness | all tools | Broad correctness and regression reviewer for behavior breakage, API/CLI/devex regressions, feature-gate leaks, migrations, state semantics, and cross-module side effects. Does NOT write code. |
| reviewer-doc | all tools | Documentation accuracy reviewer. Verifies docs match implementation, detects stale API docs, and audits architecture documentation drift. |
| reviewer-error | all tools | Silent failure and error handling reviewer. Detects swallowed errors, missing logging, inadequate error propagation, and audits catch blocks. |
| reviewer-perf | all tools | Performance validation specialist. Use for latency validation, benchmark execution, percentile analysis (P50/P95/P99/P99.9), or regression detection. Does NOT write code. |
| reviewer-quality | all tools | Code quality reviewer for maintainability, simplification, abstraction value, type boundary clarity, canonical helper reuse, and spaghetti-growth prevention. Does NOT write code. |
| reviewer-safety | all tools | Memory and thread safety auditor. Use for unsafe code audits, data race detection, or lock-free correctness verification. Does NOT write code. |
| reviewer-security | all tools | OWASP vulnerability reviewer. Use for auth logic, user input handling, API endpoint security review. |
| reviewer-structure | all tools | Code structure and modularity reviewer. Detects oversized files, god objects, module boundary violations, and untracked TODOs. |
| reviewer-test | all tools | Test coverage and quality reviewer. Verifies adequate test coverage, detects missing edge cases, and audits test quality. |
| rust | all tools | Rust engineer for performance-critical systems. Use for zero-allocation hot paths, lock-free algorithms, SIMD optimization, and systems programming. |
| tpm | all tools | Technical Program Manager for analyzing roadmaps, project lifecycle, and progress. Returns recommendations only — does not modify project management tools. |
| Event | Command |
|---|---|
| SessionStart | "${CLAUDE_PLUGIN_ROOT}/skills/git/cleanup-merged/scripts/cleanup-merged.ts" session-start |