Recently updated open-source projects

Start with a few questions about priorities, voice, audiences, and campaigns, then crawl the company site and research competitors and public reputation signals. The skill maps the findings to HubSpot's Business, Customers, and Brand Kit fields, respecting form constraints, and produces a tabbed HTML paste sheet with individually copyable values. It can also export product research, ICPs, and personas for future updates.

Claude UXR Skills is a personal toolkit for UXR and ReOps work at Miro. Its skills draft research plans and interview guides, audit prototype repositories without changing them, summarize weekly activity, and recap action items from Slack messages. Outputs are intentionally drafts for editing, and the end-of-day recap requires a connected Slack MCP. The repository complements rather than replaces the team’s canonical research skills.

atomizekit turns a topic into a blog post, native drafts for seven platforms, branded image cards, and voiced short-form video, then gathers the files in a local review dashboard. It configures itself from the host project's brand assets and can track gaps in AI-answer citations. Everything remains draft-only for human posting, and video output requires tools such as FFmpeg and Chrome.

This toolkit helps teams plan and interpret user research before building. It creates research plans, screeners, and non-leading interview guides, then turns transcripts or notes into thematic or matrix-based synthesis with an explicit build, pivot, or kill decision. Local DIY use is the default with no account or network required; community, synthetic, and paid backends are optional.

Suno Craft — LV wraps KIE.ai’s asynchronous Suno API in a resumable submit, wait, status, and fetch workflow. Beyond making songs, it supports sound effects, extensions, covers, stem separation, WAV and MIDI extraction, personas, voice cloning, video, cover art, and word-level lyrics timing, plus a field guide to music craft. A cost-tier gate asks before expensive batches, the API key stays in a local file, and generated files must be downloaded before temporary URLs expire. Users remain responsible for service terms, rights, and AI disclosure.

This repository packages two Claude Code plugins for Fused. The primary agent-core plugin covers setup, infrastructure, project authoring, execution, verification, storage, and widgets; workbench preserves legacy canvas, JSON-UI, UDF, and integration skills. Both rely on the fused CLI, with the vector extra recommended for local DataFrame handling. It is aimed at developers building and operating Fused projects from Claude Code.

steam-debugger triages Steam issues on SteamOS, Bazzite, ChimeraOS, Arch-based systems, and related Linux gaming distributions. It asks for the distro, GPU, session mode, and symptom before routing to gamescope, Proton, graphics, input, audio, or display guidance. Read-only shell collectors gather structured snapshots, while fixes are labeled from official to deep and risky; the highest-risk tier requires explicit confirmation.

Project sendbox letters, handoffs, and external tracker tickets into one Markdown table where each row represents an atomic user action. The dashboard supports active and archived sections, action types, completion ownership, and a rolling archive without SaaS, SQL, or a web UI. It complements—not replaces—the underlying trackers and sendbox protocol.

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.

Use this Claude skill to revise landing pages, sales pages, e-books, case studies, product copy, and founder notes. It removes process bleed and recurring AI-style phrasing, strengthens vague passages with details found in your material, and preserves the supplied brand voice. Unsupported metrics, testimonials, or claims become visible placeholders rather than fabricated evidence, and important caveats stay intact. The result is a strong template for a human to finish, not an AI-detector bypass.

This bundle installs an orchestrator, a section-drafting skill, and five reviewer subagents for position papers, preprints, and journal-style articles. The workflow establishes primary sources and claim-to-citation mappings before drafting, then reviews argument structure, source fidelity, vocabulary consistency, and reader clarity in parallel, with citation formatting as the final gate. The skills and reviewer agents must be installed together; a skills-only installer leaves the review loop incomplete.

Before custom code is written, the skill makes an agent state the need, search the current repository, package registries, MCP servers, installed tools, and maintained open source options, then document a decision. The final verdict must choose Adopt, Extend, Compose, or Build using evidence such as fit, maintenance, documentation, licensing, and dependencies.

Readme Writer separates issues that code can verify from qualities that require editorial judgment. Its Python linter checks the heading outline, image alt text, and local links, while Claude reviews the lead, value proposition, narrative flow, and reader clarity without reducing them to a score. It returns concrete findings and small proposed edits for human approval, and is intended for repository READMEs rather than llms.txt files or long-form articles.

Translate English prose, UI strings, subtitles, fiction, or marketing copy into Hungarian while handling focus order, definite conjugation, case suffixes, register, and domain conventions. Reference files add literary and localization rules when needed, and brand names, code, URLs, and placeholders stay intact. It only targets English-to-Hungarian translation.

TOON Skill rewrites verbose coding-agent prompts into tighter specifications while preserving exact paths, constraints, and completion criteria. It encodes uniform tabular data in Token-Oriented Object Notation when that reduces size, but keeps nested or irregular data as compact JSON. The bundled Python converter can use the official CLI when available and provides a pure-Python fallback.

cockpit aggregates the sources you wire in—CLIs, MCP servers, or local files—then ranks items by urgency, convergence across surfaces, and stakes. It renders a single HTML brief with the data embedded, so the file opens from file:// or on mobile without a server. The skill is intentionally read-only: it never sends, moves, or writes anything, and the standalone renderer needs only Python 3.

Use this multi-agent toolkit to inspect Search, Performance Max, App, Display, Shopping, and Video campaigns across one or more manager accounts. It covers conversion and Google tag health, keyword and auction research, pacing, anomalies, asset quality, and placement safety, while keeping exclusions, mutations, and campaign creation behind validation and explicit approval gates. Setup requires Python, gcloud, and a Google Ads developer token, with locally enforced 24-hour sessions.

Deck Studio guides deck creation through a brief, design specification, outline, visuals, and shape-based PPTX rendering. Each step has an artifact and confirmation point, so changing a color, rewriting a slide, or cutting the page count does not require editing the final deck by hand. It works as a Claude Code skill or through a Streamlit UI and CLI, with Chinese-market examples and editable PowerPoint or Keynote output.