apple-cktool-js
Builds and troubleshoots CloudKit automation with Apple's CKTool JS packages, including `@apple/cktool.database`, `@apple/cktool.target.nodejs`, and `@apple/cktool.target.browser`. Use when Codex needs typed JavaScript or TypeScript for CloudKit schema import, export, validation, or reset; container and team discovery; record queries, creation, updates, or deletion; Node.js or browser configuration; Promise-based API calls; CI integration; token injection; or migration from `xcrun cktool` shell scripts. Use `apple-cktool` alongside this skill for the Xcode-bundled macOS CLI.
apple-cktool
Operates Apple's Xcode-bundled `cktool` command-line utility for CloudKit development automation on macOS. Use when Codex needs to run, script, explain, or troubleshoot `xcrun cktool`; manage `.ckdb` schemas with export, validate, import, or reset; discover CloudKit teams; create, query, or delete test records; configure management or user tokens in Keychain or CI; or prepare CloudKit-backed integration tests. For equivalent typed JavaScript or TypeScript automation, use `apple-cktool-js` with this skill.
audit-training-experiment
Audits an ML training experiment for reproducibility, evidence integrity, leakage, checkpoint identity, evaluation validity, and overclaimed conclusions. Use before accepting results, spending on a longer run, publishing artifacts, or handing work to another agent.
awk
>
close-out-ml-experiment
Preserves positive or negative ML experiment evidence, reconciles report discrepancies, records the canonical decision and limitations, and prepares a concise handoff. Use when stopping an experiment, rejecting a prompt/model, or freezing a result before the next iteration.
compare-models-blindly
Creates seeded blinded model-comparison packets and evaluates candidate outputs twice with reversed order to control identity and presentation bias. Use when two models or prompts produce different valid answers that require qualitative judging.
design-resumable-model-evaluation
Designs strict model benchmarks that persist per-case evidence, resume without repeating completed work, separate first-attempt behavior from remediation, and stop early only when failure is mathematically certain. Use for slow, costly, or interruptible evaluations.
gogopowermanager
Orchestrates and ships a milestone, release, issue set, or multi-workstream engineering goal through one manager and bounded parallel workers. Use when the user asks to manage a milestone, build a dependency graph, execute independent issues in parallel, coordinate threads, sessions, subagents, or worktrees, recover a stopped delivery, or invokes /gogopowermanager.
ios-design-review
>
mikrotik-routeros
Configures and automates MikroTik RouterOS routers through the terminal, command line interface, .rsc imports, /system script repository, Scheduler, Netwatch, firewall, NAT, DHCP, DNS, bridge, VLAN, routing, VPN, device-mode, backup/export, and RouterOS scripting. Use this skill whenever the user mentions MikroTik, Mikrotik, RouterOS, WinBox terminal, router CLI, router scripts, .rsc files, or configuring MikroTik services from the command line.
orbstack-cli
>-
package-private-ml-artifacts
Inventories, hashes, scans, and verifies datasets, adapters, checkpoints, logs, and evaluation evidence before an explicitly approved private upload. Use when ephemeral jobs or handoffs require durable ML artifacts without exposing credentials or unapproved data.
prepare-training-corpus
Builds and freezes deterministic train, validation, and test corpora with provenance, group-safe splits, manifests, hashes, and prompt-leakage checks. Use before teacher labeling, fine-tuning, or comparing models on generated or retrieved examples.
proxmox-admin
Administers Proxmox VE hosts, creates and manages VMs with qm, manages LXC containers with pct, configures storage, networking, clusters, and automates provisioning tasks via the Proxmox CLI.
review-ios-ui-apple-hig
Review iOS app UI against Apple Human Interface Guidelines and iOS platform conventions. Use when asked for an Apple HIG-based iOS UI, UX, SwiftUI, UIKit, screenshot, simulator, accessibility, Dynamic Type, navigation, layout, controls, color, typography, or platform-conventions review, especially when the user wants critique or prioritized findings grounded in Apple's interface guidance instead of project-specific design docs.
review-training-data-quality
Audits a candidate or labeled training corpus for distribution collapse, ambiguity, context sufficiency, hard-negative quality, abstention behavior, leakage, and stable label defensibility. Use before scaling teacher calls or starting fine-tuning.
run-hugging-face-training-smoke
Runs one bounded, private Hugging Face training smoke with immutable inputs, local preflight checks, explicit stop gates, checkpoint verification, and reproducible evidence. Use before spending cloud compute on a longer fine-tuning run.
sed
>
select-and-verify-best-checkpoint
Proves that evaluation ran at the intended cadence, the tracked metric selected the true best checkpoint, and held-out inference loaded that checkpoint. Use with Hugging Face Trainer or compatible training-state artifacts before trusting test metrics.
send-to-minion
Use this skill only when the user explicitly invokes $send-to-minion, /send-to-minion, or explicitly asks to delegate a prompt to one minion, helper, worker, child agent, spawned agent, subagent, Claude Code agent, or other single delegated agent. It sends the user's prompt to exactly one subagent using the current runtime's delegation tools, coordinates follow-up with the runtime's messaging/resume tools, verifies the result, and returns the parent agent's final answer.
swift-review
>
technical-writer
Writes and revises clear, precise, accessible technical content for any medium. Use for manuals, wikis, README files, developer and user documentation, tutorials, references, specifications, RFCs, feature requests, issues, commit messages, code comments, review comments, pull request messages, release notes, runbooks, decision records, status updates, email, Slack, and other workplace or agent-facing communication.
tmuxinator
Helps create, edit, and debug tmuxinator project configurations, set up complex tmux session layouts, and automate development environment startup with multiple windows and panes.
topydo
Manages todo.txt tasks using topydo CLI. Adds, lists, completes, prioritizes, tags, and organizes tasks with dependencies, due dates, recurrence, and projects. Use for any task management, todo lists, or when the user mentions tasks, todos, or todo.txt. Requires Python 3 and pip. Works on macOS, Linux, and Windows.
unhobble-agent-context
Audits and designs lean, coherent context across system prompts, persistent instructions, repository guidance, skills, tool descriptions, memory, and references. Use for context engineering; creating, reviewing, simplifying, or reorganizing agent context; resolving conflicting, duplicated, bloated, stale, or misplaced guidance; designing progressive disclosure; or deciding where shared information should live.
validate-structured-model-output
Validates JSON or tool-call model outputs without repair, separating parse, schema, supplied-identifier, semantic, replay, and safety failures. Use for first-attempt benchmarks, teacher-label pipelines, and bounded model tool selection.
write-a-prompt
Creates a copy-ready prompt from a rough request, notes, source material, or the current conversation using OpenAI's prompting guidance. Use when the user invokes $write-a-prompt or /write-a-prompt, types a common misspelling such as /write-a-promopt or /wite-a-prompt, asks to "write a prompt for me," asks to turn the conversation into a prompt, or wants a prompt tailored for Chat, ChatGPT Work, or Codex. It writes the prompt without performing the underlying task.
conventional-commits
Formats commit messages using the Conventional Commits specification. Use when creating commits, writing commit messages, or when the user mentions commits, git commits, or commit messages. Ensures commits follow the standard format for automated tooling, changelog generation, and semantic versioning.
obsidian-daily
Manages Obsidian Daily Notes via the official Obsidian CLI. Creates and opens daily notes, appends entries (journals, logs, tasks, links), reads daily notes and vault files, searches vault content, and handles relative dates like "yesterday", "last Friday", and "3 days ago". Requires Obsidian 1.12+ with Command line interface enabled.