Recently updated open-source projects

Snapshot every accessible Azure subscription through the signed-in az CLI, then evaluate cost waste against versioned Microsoft and FinOps Foundation rules. The tool produces a forwardable Markdown report and machine-readable findings with severity, confidence, and estimated GBP savings ranges. Its command wrapper blocks mutating Azure verbs, so it investigates opportunities but does not remediate them.

Use a WHY → WHAT → HOW review to reconstruct the project’s goal, map its architecture and trust boundaries, and inspect implementation details through that lens. Findings are ranked by leverage, so goal or architectural mismatches appear before naming and formatting nits. It is suited to AI-generated, inherited, or pre-refactor codebases, not small diffs or greenfield projects with no code yet.

Capture a thought without leaving the terminal: the plugin detects the track currently playing in Spotify and appends a timestamped entry to a local Markdown file for that song. You can recall notes for the current track or search the whole archive. It uses no Spotify API, OAuth, or cloud sync; macOS and Linux are supported, while Windows is not.

Provide a type specimen and the plugin selects a similar OFL-licensed Google Font, renders and bitmap-traces 82 target glyphs, rebuilds the font tables, renames the family, and emits an OTF plus an HTML specimen page. It requires potrace, Pillow, and fonttools, focuses on Latin glyphs, and performs no hinting pass.

This small Claude Code plugin library currently offers two manual workflows. The ETL skill stages raw research material before transformation, while Tournament dispatches multiple parallel drafters, ranks their outputs with a decisive judge, and can synthesize the strongest result; automatic model invocation is disabled so people choose when the extra process is worth using.

This Claude Code plugin intercepts assistant replies containing can’t or cannot and asks the model to name the actual constraint, list feasible alternatives, explain tradeoffs, and recommend a path forward. It is intentionally narrow—a Stop hook plus a bundled reframe skill—and may misfire on genuine safety or legal refusals.

Use this Claude Code plugin with its WXT factory to scaffold or import an extension, prepare listing copy and screenshots, configure Chrome Web Store credentials, and package or submit the release. Its validators catch store-specific issues before shipping. The workflow automates release preparation, but it does not build the extension’s actual product features for you.

Zero-Context Validation checks whether documentation is truly sufficient by giving a fresh sub-agent only the docs and a concrete implementation task. Any missing interface, file location, or contract must be reported instead of guessed, so the main agent can revise the material and repeat the test. It is designed for Claude Code or compatible agents and uses their built-in task mechanism.

Automation Blueprint asks focused questions about a manual process, scores its repetitiveness, volume, data structure, API access, and economic value, then produces a friction map, Mermaid architecture, ROI ranges, and phased implementation plan. It labels facts, estimates, and hypotheses, adds human approval gates for destructive steps, and can recommend staying manual when the score is too low.

carousel-maker turns an article, X thread, or reel into an 8–10-slide Instagram carousel. It gathers source text and visuals, designs a topic-fitting palette and story, builds a 1080×1080 HTML deck, and renders static PNG or motion MP4 files in post order. It asks for brand choices on first use and can iterate on feedback; reels require yt-dlp, while rendering uses ffmpeg and Playwright with Chrome.

buy-me-a-car combines 16 Claude Code skills into a buyer-side workflow for vehicle research, quote collection, out-the-door calculations, dealer replies, CARFAX review, insurance, and closing checks. The full flow searches multiple sites and creates Gmail drafts for review rather than sending binding messages automatically. It is a single-author alpha, and fee, incentive, and eligibility data can drift, so figures should be verified before signing.

Binharic is a terminal coding agent with a Tech-Priest persona and support for models from OpenAI, Google, Anthropic, and Ollama. It uses built-in tools, keyword-based retrieval, predefined development workflows, and MCP extensions to handle common coding tasks. The project is still in early development, so bugs and breaking changes are expected.

picocode is a compact Rust coding agent for interactive sessions, shell pipelines, automated recipes, and focused work such as CI fixes or small codemods. It supports multiple hosted model providers as well as Ollama, and includes filesystem, search, and shell tools. Destructive file operations and command execution require confirmation by default, and the agent can also be embedded as a Rust library.

geo-content-gap compares where an AI engine retrieves your pages with where it actually cites them for each tracked prompt. It lists your retrieved or cited URLs, the third-party sources being cited, and routes each prompt to no action, existing-page optimization, or new content. The output includes a prioritized backlog and optional briefs in Markdown, Google Sheets-ready TSV, and Excel fallback formats, using whatever visibility source supplies the required per-prompt data.

Ask whether one CVE affects your environment and receive an affected, not affected, or needs verification verdict with each deciding condition tied to a source. The skill builds a persistent local environment profile and checks version, configuration, vendor fixes, and attacker prerequisites. It is for triaging scanner findings rather than replacing scanners or producing compliance records, and it needs filesystem access plus live web access.

decard finds decorative card containers in shadcn, MUI, Bootstrap, Vuetify, Ant, Chakra, or custom UI and replaces them with sections, headings, and dividers. It preserves structural frames around tables, lists, code blocks, and inputs, and can collapse repeated card grids into divided lists. Use it to reduce visual wrapping without removing meaningful boundaries.

sync-mac-windows keeps the portable parts of ~/.claude aligned between Windows and macOS using a Git repository plus a OneDrive mirror. It translates paths, preserves OS-specific MCP forms, and stages only sanitized skills, agents, commands, memory notes, and shared settings while excluding tokens, sessions, caches, and machine-only files. Audit, stage, apply, and mirror scripts default to dry runs.

Describe the grade, subject, and teaching goal, and the German-language Claude skill selects an AIS.chat learning scenario, dialogue partner, or assistant template, then fills the form fields within the documented May 2026 limits. It also inserts mandatory brevity instructions to reduce verbose chatbot replies. Claude's code-execution capability must be enabled, and the finished fields are copied into AIS.chat manually.