Recently updated open-source projects

This Claude Code skill applies a compact 20-rule checklist while writing or reviewing Java, covering naming, function design, SOLID, testing, concurrency, and clean architecture. It cites violations by rule name and proposes concrete changes, supported by on-demand references and before-and-after examples for Spring Boot, plain Java, and domain code. It catches mechanical code smells but does not replace human review of architecture, security, or business logic.

Read Deeper works with books stored in NotebookLM or files uploaded directly as PDF, EPUB, DOCX, or text. Its twelve modes cover core arguments, chapter distillation, 30-day actions, permanent notes, blind spots, debate, cross-book links, recall testing, and other focused outcomes instead of a generic summary. All output is in Thai with source markers. NotebookLM use requires a configured MCP server, while uploaded files need no additional integration.

Provide a birth date, time, and place to calculate a Western or Vedic chart and produce a styled DOCX with chart graphics, placement interpretations, and practical suggestions. The project also includes modules for transits, progressions, synastry, relocation, and related readings; local use depends on Python astrology libraries, while Claude composes the interpretive text.

Claude Prompt Booster chooses among three modes from the request: rewrite an existing prompt, design one from scratch, or audit it against a ten-criterion checklist. It applies structured prompt patterns, examples, model-specific guidance, and hallucination guardrails drawn from Anthropic materials. Boost and build modes return a copy-ready prompt with rationale, while audit mode prioritizes fixes without rewriting unless the user explicitly asks.

Password Standard is a humorous Claude Code skill that generates a single password-like string of at least 247 characters containing a differential equation, a verified prime, an anti-hero narrative, and a Shyamalan-style twist. The exaggerated constraints are the point, and many password fields cannot accept the output. The project explicitly says not to use these creations for banks, important accounts, or any real security purpose.

The skill examines relevant work across static code, dependencies, secrets, infrastructure, authentication and authorization, business logic, and runtime observability. It can generate hardened code, threat models, bug-bounty reports, and compliance gap assessments mapped to OWASP, CWE, NIST CSF, ISO 27001, and CIS Controls. Each result includes a scorecard and residual risks rather than claiming the system is secure.

imap-client lets an AI agent turn inbox questions into read-only myl commands for searching messages, viewing content, and downloading mail or attachments. Credentials stay in environment injection or a permission-restricted file rather than generated commands. It supports Gmail, Yandex, Mail.ru, and many discoverable IMAP providers, but it cannot send, move, delete, or label messages.

Give Claude a claim, contract, or argument and it responds as a rigorous opponent, identifying the most dangerous weaknesses, unanswered questions, and plausible worst-case outcome. The skill is written primarily for Russian-language legal work but can also challenge business ideas and general reasoning. Its output is preparation material, not a substitute for qualified legal advice.

Give Claude a topic, audience, goal, and any source material, and the skill develops a slide-by-slide narrative with speaker notes, layout choices, and visual recommendations. It returns an editable Markdown deck structure for business, sales, training, technical, and other presentations rather than producing a finished PowerPoint file.

This skill adds live web search to OpenClaw for time-sensitive questions such as news, policy changes, and market developments. It relies on curl plus an ANSPIRE_API_KEY and can be installed through ClawHub or copied into a workspace. Its instructions also require explicit fallback behavior when a search fails instead of presenting stale information as current.

Give it a media URL or a short description, and it creates a ready-to-use agenda for book clubs, watch parties, or dinner conversations. Questions are organized around an icebreaker, central tensions, deeper follow-ups, personal connections, and a closing prompt, adding structure without turning the session into a script.

Use this skill as an implementation playbook for a single-file CRM, a drag-and-drop automation editor, trigger-condition-action workflows, a session gate, and an animated brand landing page. It includes architecture references and starter files derived from a working project. The prescribed stack is fairly specific, and the user experience is tuned for Latin American clients with Spanish interface copy.

context-kit interviews you to create role and project packs, then distills durable facts, decisions, open questions, and explicitly rejected options from ongoing conversations. The source stays as plain Markdown on your machine and can be exported for ChatGPT custom instructions, Claude Projects, or raw reuse elsewhere, without an account, server, or synchronization service.

Content Researcher uses Apify data from roughly 100 public Instagram or TikTok posts to produce a structured Word report covering performance, content pillars, competitors, and the next ten videos to make. It requires Claude Pro or Team plus Apify, and its per-video judgments come from captions and thumbnail metadata rather than retention data or frame-level analysis.

Design Makeover turns a vague request such as “redesign my app” into a guided workflow: it scopes the job, asks a brief set of questions, audits the current interface, proposes distinct directions, implements the selected one, and reviews its own work. It adapts its approach to dashboards, marketing pages, editorial views, or mobile apps, with Playwright recommended for autonomous browser checks.

Tell Claude Code which supported provider you want to use and this skill creates the appropriate shell or PowerShell launcher, adds an alias, and gives you a short command for starting Claude Code against that backend. It covers GLM, DeepSeek, Qwen, and Kimi configurations across common shells. You still need a valid provider API key, and a new terminal session is usually required before the alias works.

Use this Claude Code skill to create the right robots.txt, ai.txt, llms.txt, or TDM opt-out file for a website's actual policy. Its audit mode checks existing files for outdated AI user agents, crawler inheritance traps, and mismatches between training and search access, while distinguishing unrelated specifications that share the ai.txt name.

ccthread lists, finds, searches, and renders local Claude Code sessions so users can recover configuration values, past decisions, rollback reasons, or weekly work summaries. It works as an auto-invoked Claude Code plugin or a standalone CLI, filtering out most bookkeeping while preserving readable messages and attached tool results.