Shape agent capabilities into your workflow.

Use sprint-orchestrator to keep one Claude Code chat focused on planning and review while separate sprint chats implement features in isolated Git worktrees. Each sprint can open a PR and preserve decisions and recurring bug lessons in repository state, reducing context growth across long projects. It is aimed at ongoing, multi-feature development rather than one-off scripts or tiny fixes.

Give Claude a brief, prototype, deck, or URL, and the skill guides it through scripting, a beat-by-beat motion board, design review, a roughly 15-second hook render, and the final 60–90 second film. It is a workflow rather than a standalone app and requires a skills-enabled Claude environment, Node.js 22, FFmpeg, a renderer such as Remotion or HyperFrames, and ElevenLabs access for voiceover.

This skill structures a product’s pre-launch legal and privacy review around source documents, user paths, checklists, prioritization, and a 5×5 inherent-versus-residual risk matrix. It produces a summary and detailed review with Tier 1–3 workstreams, and can optionally publish to Confluence; it is a working aid, not legal advice or sign-off.

expo56-skill packages Expo SDK 56 guidance into twenty topic-based documents that Claude Code loads as needed. It supports new app development and SDK 55 migrations across routing, native modules, file handling, media, authentication, tooling, and EAS workflows, helping avoid outdated or invented APIs. The material is a May 22, 2026 beta snapshot rather than a live mirror, so release-critical versions and interfaces still require verification against official Expo sources.

Use this Claude Code skill to plan and carry out an eight-stage paper workflow, including topic refinement, drafting, polishing, first-pass review, defense preparation, and submission. It handles general writing work directly, then recommends and hands off to specialized skills for tasks such as deep literature research, framework figures, or institution-specific document formatting. Its built-in checks emphasize source-grounded claims, hypothesis discipline, and readable prose.

After one-time project setup, the skill creates a shared task board, a file-lock registry, role-specific instructions, and a planner approval gate. Multiple Claude Code terminals claim tasks and files before editing, then stop for testing and review before committing. It supports solo work through teams of five or more terminals, but its locks are a coordination protocol that every session must follow, not operating-system enforcement.

JDfit compares a job description with a DOCX resume, scores match, quantified impact, structure, language, and ATS readiness, then edits the original document while preserving its layout. Every change receives a native Word comment describing the original text, revision, and rationale, and the tool saves both a backup and an annotated version. Missing evidence is flagged for the candidate instead of being invented.

Use Elevate to review a web, macOS, or iOS codebase across performance, functionality, stability, reliability, security, usability, and visual polish. Its bounded report ties recommendations to concrete files and verification steps, then identifies one change to ship first. The skill produces an improvement plan only and does not edit the codebase.

Begin a focus block by defining the task, duration, finish line, and current energy, then optionally enable Do Not Disturb, close distracting apps, and start music. The skill keeps the goal visible without conversational interruptions and records the outcome, next step, and reflection in Markdown for later weekly or monthly review.

Ratatoskr changes how an AI coding agent talks to the user, keeping product changes in plain language instead of code, jargon, or invented effort estimates. It names defaults, asks focused questions when intent is unclear, and requires a clear confirmation before permanent actions such as deleting data. It does not control the development method or provide memory between chats.

Stop Slop v3 guides a model through a derivation step—reader profile, register, scene anchor, texture, and optional voice sample—before drafting. Positive sentence-level constraints shape the prose, and an optional final audit targets visible AI patterns after generation. The diagnostic catalog is kept out of the drafting context.

memphis-html-ppt turns local Markdown or a web page into a self-contained HTML slide deck with bold color, geometric accents, and editorial layouts. It extracts and cleans the source, plans slide content, renders the preview, and checks structure, content quality, and template fit before sharing. The output opens in a browser, while complex article structure and long sections may still be compressed.

Use this agent skill to compare API spending, rented-GPU self-hosting, and fine-tuning ROI from workload details in messages or project files. It fetches current prices, runs deterministic VRAM and cost calculations in a standard-library Python engine, and produces a sourced report. Treat the result as directional: default estimates exclude operations and engineering, assume one replica, and size model weights without full KV-cache demand.

These Claude Code skills gather material from repositories, Git history, or pull requests, turn it into stakeholder-friendly HTML, and publish it through the Uselink CLI. They cover repository overviews, PR digests, changelogs, API documentation, onboarding guides, reports, and planning specs. Publishing requires a Uselink account, an API key, and a configured CLI.

Use the skill to review French text for stacked short sentences, chained present participles, empty intensifiers, anglicisms, false contrasts, and formulaic phrasing. It favors more naturally connected French syntax and can score a draft across five writing dimensions before revising it. The tool focuses on style and typography, not detailed spelling or grammar correction, factual accuracy, argument quality, or specialized literary and legal registers.

This Claude Code skill structures substantial engineering work around understanding, specifying contracts, pre-mortems, small changes, and behavior-focused verification. It adds 30 red-flag checks, software and security thinking layers, and situation-specific playbooks grounded in ten engineering books. The project states that real-world use remains unproven at scale.

Give Podcast Workflow a topic and it researches angles, writes a conversational script, pauses for your approval, and synthesizes the finished audio through Xiaomi MiMo TTS. You can use built-in Chinese voices or clone a voice from a 30–60 second sample, with paragraph-level emotion cues. It requires Claude Code and a MiMo API key.

Configure each price, version, rule, or deadline with its owning source page and the local files that cite it. fact-drift saves a baseline, extracts the current value, and shows the old value, new value, source quote, and stale matches before changing anything. Updates require approval, while login-only pages, heavy JavaScript sites, and PDFs need a manual or separate browser-based check.