Xopoko
UserReady-to-install skills and plugins for Codex, Claude Code, and AI coding agents: practical workflows for app delivery, architecture, research, design, and agent tooling.
Categories
Indexed Skills (93)
capability-portfolio-architect
Reshape skill/plugin portfolios when routing overlaps, capabilities duplicate or disappear, or split/merge/move/delete/router/reference/script boundaries need decisions. Use capability-auditor for single-artifact quality.
architecture-conformance
Architecture conformance compares intended rules with implementation for dependencies, ADRs, ownership constraints, drift, erosion, recovered models, and classifications.
architecture-fitness-functions
Architecture fitness functions turn intended boundaries into dependency rules, cycle checks, ownership gates, ADR conformance, runtime/resilience checks, and staged CI enforcement.
architecture-ownership-topology
Architecture ownership analysis maps CODEOWNERS/OWNERS coverage, ownerless modules, cross-owned dependencies, review paths, and governance risk without inferring team health.
architecture-refactoring-strategy
Architecture refactoring strategy plans and executes incremental code-boundary changes with characterization tests, per-slice proof, fitness functions, rollback, and before/after evidence; excludes routine cleanup and agent-runtime design.
codebase-architecture-audit
Codebase architecture audits recover actual modules, dependencies, domain seams, runtime coupling, ownership, quality attributes, tests, docs, and risks before structural code changes.
app-icon-studio
Apple app icons: create, generate, evaluate, export, install, or debug iOS AppIcon.appiconset and macOS .icns assets for small-size clarity.
apple-dev-research
Apple developer articles: search Swift, SwiftUI, Xcode, iOS, and macOS community blogs, tutorials, and write-ups, not official docs.
apple-firmware-inspector
Apple firmware: inspect and reverse-engineer IPSWs, kernelcaches, dyld shared caches, private headers, entitlements, Mach-O binaries, KEXTs, and security internals with `ipsw`.
appstore-ads-operator
Apple Ads campaigns: inspect and manage separate auth, orgs, ad groups, creatives, keywords, reports, and API calls; approve live mutations first.
appstore-archive-uploader
App Store IPA/PKG archives: set version/build numbers, archive, export, upload, or publish with `asc xcode` before TestFlight/App Store submission.
appstore-aso-auditor
App Store ASO audit: analyze canonical `./metadata` offline after `asc metadata pull`; add Astro MCP keyword gaps and Apple app-tag context when available.
appstore-connect-cli
App Store Connect commands: discover and run `asc` CLI auth, schemas, canonical verbs, pagination, output, Apple Ads, and timeouts.
appstore-notary-runner
macOS Developer ID notarization commands for xcodebuild export, `asc notarization` submit/status/log, and stapling. Excludes packaging-readiness reviews and signing-only diagnosis.
appstore-record-creator
App Store Connect New App creation via visible browser automation after bundle-ID registration for the API-less web form; never store cookies or auto-retry Create.
appstore-release-director
iOS App Store release orchestration from a local repo through signing, metadata, privacy, screenshots, upload, TestFlight, review submission/resubmission, blocker triage, and release evidence.
appstore-release-notes-writer
App Store What's New notes and promotional text from git history, bullets, or prose, with optional localization. Excludes full-listing translation, metadata sync, and subscription/IAP names.
appstore-release-planner
App Store release go/no-go planning for readiness, first-submission blockers, sequencing, and stage-versus-submit decisions. Routes execution to focused skills; review commands belong to appstore-review-readiness.
appstore-revenuecat-sync
App Store Connect and RevenueCat subscription/IAP reconciliation with `asc` and RevenueCat MCP for catalog bootstrap, drift audits, deterministic product/entitlement/offering/package mapping, and no deletions.
appstore-review-readiness
App Store review-readiness execution with current `asc` commands to validate, stage, submit, monitor, cancel, or repair blockers after go/no-go planning. Excludes release strategy; appstore-release-planner owns it.
appstore-screenshot-pipeline
iOS App Store screenshot automation with xcodebuild/simctl capture, AXe plans, Koubou framing, review artifacts, and `asc` upload.
appstore-screenshot-studio
App Store marketing screenshot creation and revision to translate, scrape, crop, and validate `.appstore-screenshots` workspaces. Excludes general image generation.
appstore-signing-setup
App Store signing asset setup with `asc` for bundle IDs, capabilities, certificates, profiles, local install, rotation, and encrypted team sync.
appstore-workflow-runner
Manage `.asc/workflow.json` automations; define, validate, run, resume, and audit trusted repo-local release/TestFlight flows and step outputs with `asc workflow`.
build-swift-apps
Route broad or ambiguous Swift and Apple-platform work to a focused skill; this router does not implement domain work. Covers iOS, macOS, SwiftUI, Xcode, Simulator, App Store Connect, Tuist, SwiftPM, signing, profiling, and Apple research.
ios-ettrace-profiler
iOS ETTrace Simulator profiles: capture and interpret symbolicated startup, scrolling, navigation, rendering, CPU hotspots, and before/after evidence.
ios-intents-architect
Design and implement iOS App Intents, AppEntity, EntityQuery, and App Shortcuts for Siri, Spotlight, widgets, controls, Shortcuts, and app handoff routes.
ios-memgraph-inspector
iOS memgraph leak analysis: capture, inspect, compare, and prove memory leaks with Apple's `leaks` tool, retain-cycle evidence, and before/after checks.
ios-simulator-browser
Mirror iOS Simulator runs in the Codex browser for interaction, visible proof, and hot-reloaded SwiftUI previews from importable Swift packages; exclude headless or log-only debugging.
ios-simulator-debugger
Debug iOS Simulator apps with XcodeBuildMCP for build, run, launch, UI inspection, interaction, screenshots, and logs; route user-visible mirrors and SwiftUI previews to `ios-simulator-browser`.
ios-swiftui-architect
iOS SwiftUI views and components: build or refactor navigation, state ownership, async UI, sheets, previews, and responsive layouts; exclude UIKit-only and macOS work.
macos-appkit-bridge
macOS SwiftUI-AppKit bridges: implement NSViewRepresentable, NSViewControllerRepresentable, NSWindow, panels, responder chains, or menus only where pure SwiftUI cannot model the behavior.
macos-liquid-glass-designer
macOS SwiftUI Liquid Glass UI: modernize or review system materials, toolbars, search, controls, and custom glass; prefer native structure over hand-built chrome.
macos-runtime-debugger
macOS app runtimes: build, launch, and debug Xcode or SwiftPM GUI/CLI targets with shell-first workflows; diagnose compiler, linker, startup, log, and telemetry failures; exclude iOS Simulator work.
async-state-consistency
Asynchronous state consistency: cache races, subscriber notifications, memoized/coalesced loads, replay, one-shot reads, invalidation, stale results. Excludes UI-only display, deployment topology, distributed consensus, unrelated flakiness.
architecture-intelligence
Software architecture routing covers code boundaries, dependencies, AI-assisted refactoring, runtime topology, async state, ownership, ADRs, fitness functions, and conformance; excludes agent-runtime design, UI/UX, and routine cleanup.
architecture-runtime-topology
Runtime architecture analysis maps services, app/CLI/background flows, deployment/IaC, integrations, observability, resilience, and coupling without claiming production truth.
agent-harness-engineering
Design agent harnesses for typed loops, runtime reconfiguration, provider/tool/module hot swap, state, policy, cancellation, recovery, and delegation. Excludes prompt-only, generic architecture, vendor CLI/config, and evaluation-only work.
agent-harness-evaluation
Evaluate agent harness reliability through replay, restart, cancellation, runtime reconfiguration, concurrent generations, rollback, context pressure, and release gates. Excludes generic tests, surveys, and prompt/model-only benchmarks.
agent-harness
Agent runtimes and human credential handoff: route harnesses, Codex/Claude operations, 1Password or native prompts, sessions, deferred completion, and scheduler proof. Excludes capability authoring and generic app architecture.
claude-agent-worktrees
Claude Code sessions coordinate background agents, `claude agents --json`, dispatched-session defaults, worktrees, tmux/iTerm panes, resume/continue/from-pr/fork-session, names and IDs, remote control, and cloud ultrareview runs.
claude-code
Claude Code CLI routes local inspection, interactive or print-mode automation, plugin and MCP lifecycle, diagnostics, hooks/settings, background agents, worktrees, sessions, remote control, and ultrareview.
claude-doctor-debugger
Claude Code diagnostics isolate install, update, auth, configuration health, broken customizations, safe/bare modes, debug logs, auto-mode, keychain/API-key boundaries, IDE/Chrome integration, doctor warnings, and setup-token failures.
claude-hooks-settings
Claude Code settings and hooks are created, audited, or debugged across CLAUDE.md/rules loading, custom agents, tool permissions, output styles, workflows, setting sources, safe/bare modes, and plugin customizations.
claude-plugin-mcp-manager
Claude Code plugins and MCP servers are managed across marketplaces, validation, token cost, install/update/remove/prune, session-only sources, MCP approvals, strict config, transports, headers, OAuth, and lifecycle failures.
claude-print-automation
Claude Code print-mode runs are prepared or debugged with `claude --print`, text/JSON/stream-json I/O, JSON Schema, budgets, fallback models, no persistence, prompt suggestions, and trusted-directory CI.
codex-cli
Route Codex CLI operations across diagnostics, exec/review automation, deferred completion, task supervision, plugin/MCP management, rollout forensics, cross-task corpora, and app environments.
codex-deferred-completion
Complete long-running executable work through an existing atomic JSON terminal receipt, avoiding repeated native-session or remote-status polling and wasted model turns.
codex-doctor-debugger
Diagnose current Codex CLI health and model-visible skill catalogs: metadata truncation/omission, install, config, auth, sandbox, prompt, app-server, remote-control, and runtime failures. For what an existing task saw, use codex-log-reader.
codex-environments
Manage Codex app project environments and actions in `.codex/environments/environment.toml`, including Run/Test/Preview buttons, startup commands, launchers, environment variables, and repeatable local commands.
codex-exec-automation
Automate non-interactive Codex CLI runs with `codex exec`, resume, review, JSONL events, output schemas, last-message files, cwd/profile/config, sandbox/approval modes, and CI.
codex-log-reader
Inspect Codex rollout JSONL by CODEX_THREAD_ID, cwd, query, issue, project, lineage, malformed or large logs, permission concerns, and "what happened in this task?" forensics.
codex-plugin-mcp-manager
Manage Codex plugins, local marketplaces, cache visibility, and MCP server list/get/add/remove/login/logout, bearer-token environment bindings, and plugin/MCP installation failures.
codex-thread-supervisor
Supervise live Codex tasks by ID only on an explicit user monitor/supervise request, using cursor waits, attention/completion gates, bounded claims, checkpoints, and authorized handoffs. Excludes rollout forensics, current-turn subagents, and external jobs.
scheduled-automation-runtime
Local scheduler jobs need proof when launchd, systemd, cron, or Windows Task Scheduler differ from manual runs or lack runtime proof. Not for vendor CLI command construction, architecture inventory, cloud schedulers, or job business logic.
credential-handoff
Credential prompts and 1Password CLI: route task-scoped secrets from a human or vault into a target process without exposing values to model context, chat, logs, arguments, or files. Excludes account administration.
agent-guidance-factory
Author repository agent guidance such as AGENTS.md, CLAUDE.md, and scoped rules for load order, nested instructions, migration, audits, or stale-doc cleanup. Excludes ordinary human docs unless agents consume them.
capability-auditor
Audit capability artifacts for static safety, evidence coverage, duplication, context cost, prompt contracts, dependencies, and install risk. Use capability-evaluation for behavioral baselines and portfolio architect for structural change.
capability-evaluation
Evaluate skill, plugin, agent-guidance, or trigger-metadata artifacts against an explicit baseline and representative behavior cases, producing evidence and an adoption decision. Excludes harness reliability and static source/safety audits.
capability-reality-repair
Repair stale or false skill/script/plugin/MCP contracts when commands, schemas, paths, outputs, dependencies, install state, connector guidance, validators, or docs disagree with live reality.
capability-synthesizer
Synthesize or strengthen well-vetted agent skills/plugins from broad public sources and local or user-provided candidates, with evidence-backed comparison and adoption/rejection.
capability-workbench
Route artifact-first agent capability work: frame behavior gaps; choose, evaluate, author, validate, govern, and explicitly activate skill, plugin, guidance, or trigger artifacts. Use Agent Harness for runtime and harness-level evaluation.
plugin-factory
Build or update marketplace-backed agent plugins with manifests, skill bundles, local marketplace entries, packaging, validation, optional install/cache gates, runtime-discovery status, and Codex deeplinks.
skill-factory
Create or refactor portable agent skills across SKILL.md bodies, progressive disclosure, scripts/references/assets, packaging, and validation. For name/description-only routing work, use skill-trigger-metadata first.
skill-installer-vetter
Find, compare, vet, install, or update agent skills from catalogs, GitHub, local folders, or user references with provenance, safety, dependency, capability, and destination checks.
skill-trigger-metadata
Optimize skill names/descriptions for reliable routing under catalog truncation, omission, and cross-runtime pressure. For instruction, resource, packaging, or installation changes, continue with skill-factory or plugin-factory.
context-density
Agent context audits measure token cost, provenance, compression, typed-state drift, trigger overlap, and prompt/output contracts across AGENTS.md, prompts, skills, plugins, MCP/tool schemas, current-state artifacts, and agent handoffs.
design-intelligence
Product, UX, and technical visual-explanation routing covers framing, information architecture, interaction, usability, accessibility, compact diagrams, and design-system governance; excludes Figma, CSS, automation, and assets.
design-system-governance
Design-system governance defines reusable patterns, contribution rules, accessibility evidence, ownership, maturity, adoption, and drift; excludes CSS, Figma libraries, and token tooling unless requested.
interaction-design
Interaction design shapes task flows, affordances, feedback, state coverage, error prevention, recovery, undo, progressive disclosure, input burden, and keyboard/touch behavior.
interface-architecture
Interface architecture structures navigation, taxonomy, labels, content models, screen priority, findability, search/browse, and information hierarchy.
product-framing
Product framing clarifies user needs, Jobs-to-be-Done, outcomes, strategy, assumptions, opportunity-solution trees, HEART/GSM metrics, and discovery before interface design.
usability-accessibility-review
Usability and accessibility review audits screens, flows, or specs for heuristics, cognitive-walkthrough risks, WCAG/APG/COGA concerns, inclusive design, ethical UX, and dark patterns.
visual-communication
Visual communication audits UI screenshots, golden images, and visual diffs for hierarchy, readability, contrast, capture state, and test-harness artifacts. Do not use for screenshot generation/export, CSS, Figma, or styling.
visual-explanation
Visual explanations for coding and technical work: choose the smallest useful table, tree, pseudocode, call flow, state view, structural diff, or wireframe. Not for UI screenshot audits, data charts, image assets, or architecture recovery.
code-maintenance-audit
Changed-code audits find dead or unused symbols, stale leftovers, duplicate logic, obsolete tests, and consolidation before completion. Not for speculative architecture, style churn, or performance optimization unless explicitly requested.
provisioning-missing-tools
Missing toolchains are provisioned when absent, outdated, weak, or misconfigured commands, SDKs, runtimes, package managers, drivers, CLIs, simulators, emulators, or test/profiling utilities block or downgrade end-to-end work.
codex-task-corpus
Build bounded cross-session Codex task corpora for last-N chat audits, recurring-work retrospectives, context recovery, and capability-gap evidence. Excludes single-rollout forensics, live supervision, and automatic skill edits.
macos-notarization-packager
macOS distribution artifacts: inspect Developer ID archives, app bundles, hardened runtime, nested signing, and notarization readiness; exclude local signing-only diagnosis and direct `asc notarization` execution.
architecture-decisions
Architecture decisions record structural tradeoffs, consequences, reversibility, ownership, validation, and ADR revisit triggers; skip local choices.
appstore-build-monitor
App Store builds: track processing, find latest builds and next numbers, wait on uploads, or safely expire old builds with `asc`.
appstore-crash-insights
TestFlight crash reports: triage crashes, beta feedback, hangs, disk writes, launches, and performance diagnostics with `asc`.
appstore-id-resolver
App Store Connect IDs: resolve apps, builds, versions, groups, testers, and review submissions from names with deterministic `asc` lookups.
appstore-metadata-localizer
App Store listing text: translate and market-adapt descriptions, keywords, What's New, names, subtitles, and privacy text across locales. Excludes non-translation edits, standalone release notes, and IAP/subscription names.
appstore-metadata-sync
App Store metadata JSON: edit, validate, push, or sync canonical `./metadata`, plus legacy fastlane migration via `asc migrate`. Excludes translation-first work, standalone release notes, and IAP/subscription names.
appstore-pricing-planner
App Store subscription and IAP pricing by territory with `asc`, including price points, PPP/localized CSV imports, availability, summaries, and schedules; mutating actions require confirmation.
appstore-screenshot-validator
App Store screenshot validation and upload with live `asc` size data and macOS `sips` to resize, strip alpha, and color-convert copies.
appstore-subscription-localizer
App Store subscription localization: create or update localized display names and descriptions for groups, subscriptions, and IAPs with `asc`; exclude app listing metadata, release notes, keywords, screenshots, and pricing.
appstore-testflight-coordinator
Coordinate TestFlight beta distribution, groups, testers, and What to Test notes with `asc` for beta rollouts.
appstore-wall-publisher
Submit or update Wall of Apps entries in the App-Store-Connect-CLI repository with `asc apps wall submit`; match wall submission, addition, or update requests.
ios-liquid-glass-designer
Implement, refactor, or review iOS 26+ SwiftUI Liquid Glass with native `glassEffect`, `GlassEffectContainer`, button styles, availability gates, and non-glass fallbacks.
ios-rocketsim-operator
RocketSim iOS Simulator UI: inspect and control accessibility state, gestures, typing, hardware buttons, and CLI automation.
macos-signing-inspector
Use when an existing macOS app or binary needs code-signing, entitlement, hardened runtime, sandbox, Gatekeeper, or trust-policy diagnosis. Not for full distribution packaging or running notarization submissions.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.