Recently updated open-source projects

AI-SEO reviews a URL for factors that may affect citation by ChatGPT, Perplexity, Claude, and related search experiences. Its Claude Code commands cover citability, crawler access, llms.txt, structured data, content quality, technical SEO, and Markdown reporting, turning the audit into scores and concrete actions.

Use the single-file bbb script to list, inspect, comment on, approve, decline, merge, and create Bitbucket Cloud pull requests. It detects the workspace and repository from Git remotes and can install agent guidance for Claude Code, Codex, Cursor, and others. The tool is Bitbucket Cloud-only, and its sourced environment file must contain plain values.

geo-markdown-mirror identifies important pages in a website or documentation site and creates predictable Markdown mirrors under an /llms/ folder. It flattens JavaScript widgets such as pricing sliders, tabs, and accordions into readable tables, generates an llms.txt index, and can add alternate links or ask-AI buttons. The skill supports Next.js, Docusaurus, Vite, and static HTML, with a regeneration script to keep mirrors aligned as pages change.

Use the skill to draft or review buttons, errors, empty states, forms, and complete user flows. It diagnoses the most important problems before rewriting, follows a reusable voice profile, and checks accessibility, localization readiness, and manipulative choice framing. Legal language, financial disclosures, consent wording, and verbatim quotations are flagged with suggestions rather than silently changed.

git-finder searches for repositories that fit a stated project need, fetches the candidates, and checks their license, maintenance activity, and actual architecture before recommending three or four foundations. The report includes a primary choice, a fallback, and a quickstart showing installation and real API usage. It never invents names, URLs, statistics, or structure, treats permissive licensing as a hard gate, and does not write application code unless you ask after the validation.

Aircraft Research OSINT Skill turns a US N-number, ADS-B screenshot, or registry excerpt into a layered ownership dossier. It combines registration history, entity records, sanctions and public-record checks, and flight-pattern analysis, then states when a principal cannot be identified. The workflow relies on web search, focuses on US registrations, and treats missing indexed data as a valid result rather than inventing an owner.

METAR & TAF Decoder parses aviation weather reports field by field, covering US/FAA, international, and military conventions. It resolves the ICAO station, explains wind, visibility, runway range, weather, cloud layers, temperature, pressure, remarks, and forecast changes, then converts the result to local time and readable units. The default output combines a plain-language decode with a Discord-ready block; the timezone and temperature preferences are configurable, and ACARS or VDL2 are not supported.

Frequent-Flyer Advisor asks three to six questions about a trip, ranks three airline-mile options, and produces a 1080×1440 shareable savings card plus a full HTML strategy report. Its bundled knowledge base covers 53 programs with dated sources, and savings are calculated against real cash fares while the cents-per-point value stays internal. The generator runs offline with no third-party dependencies, but program rules and prices should be checked against official pages before use.

iOS Motion Patterns Index catalogs runnable Swift animation examples for menus, transitions, indicators, alerts, tables, collection views, and UI controls. Its dispatch file gives a few representative links and sends the agent to fuller category lists only when needed. The project intentionally complements animation how-to skills rather than duplicating them: use it when you need working code for a swipe-to-delete card or custom transition, not general theory.

Design Systems Index helps an agent find mature systems to study or adopt, covering major company libraries, design-token resources, pattern libraries, and React or React Native component collections. The main file dispatches to a few picks and category references rather than teaching token architecture or duplicating a local design-system method. Its links are curated from two community lists, so they are a starting point for selection, not a substitute for checking the current source.

Design Tools Index points agents to external tools across color, graphic assets, UI and wireframing, prototyping, animation, handoff, accessibility, research, collaboration, and design systems. Its dispatch file shows a few top picks and loads the complete category lists only when needed. This is navigation rather than instruction, curated from the Awesome Design Tools collection; the original list remains broader and may be updated more often.

Design Assets Index gives an agent a navigable catalog of external sources for stock photos, icons, fonts, color palettes, mockups, UI kits, and templates. The main skill offers a few picks per category and points to fuller Markdown lists that load only when needed. It is deliberately a link-and-note resource index rather than a design tutorial, and users still need to check each source’s availability and license.

human-review-site is a recipe and kit for propose-before-apply workflows. It serves a bespoke HTML page privately on localhost or a Tailscale tailnet, lets a person review many independent decisions, and exports one standard JSON envelope through the clipboard so the agent can apply only accepted items. Decisions persist in localStorage and the agent must test transport and rendering before presenting the URL. The examples use plain HTML, CSS, and JavaScript without a build step.

hardEningCLOUD_byDM20911 turns cloud inventory, CIS checks, and approval-gated remediation into one auditable hardening workflow. It records evidence for each finding, compares posture before and after changes, and applies anti-lockout checks before any write. The same method is packaged for Claude, Gemini, Codex, and local models; use it only on accounts or client environments you are authorized to assess.

SE Ranking Analyst resolves the tracked project and uses dates that actually exist before comparing rankings, AI-search mentions, backlinks, and site-check findings. It reports current-minus-baseline changes in words, distinguishes real no-data periods from guesses, and ends with prioritized actions and data notes. The skill works through the SE Ranking MCP connector or API and consumes the user’s account credits; it is useful for agencies, in-house teams, and consultants who need repeatable client reports.

This integration pack exposes the GhostSwap Partners API through MCP, OpenAPI, a Claude Skill, and configuration files for several coding assistants and agent frameworks. Developers can let an agent list currencies, quote swaps, create transactions, and poll status, while real swaps require an approved partner account and server-side credentials; the repository wraps the hosted API rather than running the exchange itself.

At the end of a meaningful conversation, the skill scans the full session for durable decisions, preferences, pitfalls, and open questions rather than producing a generic transcript summary. It presents a short list for editing and writes only after explicit confirmation, using KnowMine, another MCP-compatible notes tool, or clean Markdown for manual storage.

Spartan reduces filler and repetition in Claude’s incoming material, outgoing answers, or both. Its lite, full, and ultra levels move from shorter sentences to denser fragments and abbreviations, while keeping code blocks, technical terms, error messages, numbers, and security warnings intact. Clarity takes priority again for destructive actions and other safety-sensitive steps. The skill works in Claude.ai, Cowork, and Claude Code and is controlled through commands such as spartan in, spartan out, and spartan status.