Find the open-source solution that fits.

Prism Scanner applies static detection rules to agent skills, plugin folders, MCP servers, and Python projects, reporting the matched file, rule, reason, and location. It is intended as a pre-install review aid for third-party tools and can run as a Windows application or Python package. A match is a prompt for human inspection, not proof that a file is malicious.

The Xiaolu House CLI lets people or AI agents search resale homes, rentals, new developments, communities, transaction records, and schools from a terminal. Results can be filtered by city, area, rooms, price, size, direction, and other housing attributes. A Xiaolu House API key and default city must be configured before querying.

This Claude Code skill requires suspected bugs to be checked against tests, callers, and concrete execution traces before they are reported. Its strongest coverage is a 20-section C and C++ checklist with four severity buckets, while reference guides also cover Rust, Go, Python, Java, frontend frameworks, architecture, performance, and security.

This skill gives data engineers ten practical rules for writing or reviewing Hive, Impala, and Spark ETL with an AI coding assistant. It targets failures that may not raise errors, such as shifted columns, lost rows, unsafe joins, runaway aggregation, and stale metadata, and separates analysis behavior from coding behavior.

This Claude Code skill reviews Markdown-based agent definitions across 11 areas, including scope boundaries, evidence requirements, tool discipline, failure handling, security, and model fit. It returns PASS, WARN, or FAIL findings with exact line references and concrete fixes, and can be used with agent formats from Claude Code, Cursor, Codex, and similar tools.

Adaptive Review inspects free signals such as changed lines, file count, directory spread, and security keywords before choosing a fast, medium, or deep review path. Its thresholds are editable for different repositories, and an OpenClaw edition can route to OpenAI-compatible local models such as Ollama or vLLM instead of Claude tiers.

This skill creates architecture diagrams, flowcharts, CI/CD pipelines, dependency maps, or edits to an existing `.excalidraw` file without flooding the main conversation with diagram JSON. Isolated agents generate the drawing, then local scripts validate and render PNG or SVG output, with optional animated SVG export. It requires Claude Code, Node.js 20 or newer, and npm.

enggenie packages 14 role-based skills that move work from a vague request to a specification, phased plan, test-driven implementation, code and design review, evidence-based QA, and PR preparation. The stages leave structured artifacts for handoffs and work across Claude Code, Cursor, Copilot CLI, Gemini CLI, OpenCode, and other skill-compatible assistants.

Use this Claude Code skill when a life or technical decision is being driven by emotion, anchoring, narrow framing, or a repeated debugging hypothesis. It follows a fixed intervention: name the bias, replace intuition with probabilities or base rates, surface an overlooked option, calculate expected value when useful, and end with one time-bound next action. It stays silent when no intervention is needed.

Ask Claude to critique a B2B landing page, PLG motion, product-led sales strategy, pricing package, growth team, or ideal customer profile using frameworks drawn from Leah Tharin’s public work. The skill is intentionally direct and opinionated, focusing on what is broken and what to change. It is a fan-made project and is not endorsed by or affiliated with Leah Tharin.

Provide a Bilibili, YouTube, podcast, WeChat, Zhihu, or long-form article link—or a local audio or video file—and turn the material into a formatted ebook. It is designed for knowledge workers organizing saved content, students and researchers processing lectures, and creators repurposing long-form material into something easier to read and reuse.

Coordinate discovery, design, architecture, implementation, testing, release, and operations with schema-validated artifacts and script-enforced transitions. Implementation and test gates run declared verification commands rather than trusting filled-in status files, while upstream revisions mark downstream work stale. Key risk gates require confirmation, and the implementation phase depends on the separate harness-plan skill.

Run the skill at the end of a session to capture completed work, decisions, changed files, build and test status, blockers, and next steps in a dated Markdown document. The result can restart a later Claude session, serve as an end-of-day report, preserve a development log, or let a colleague continue asynchronously without a handoff meeting.

Configure a context map for your projects, then let the skill turn each opening request into a setup checklist of files to load, skills to enable, and MCP connectors to enable or disable. When the request is unclear, it asks a small number of focused questions. A mid-session check can remove deadweight or add newly relevant tools as the work changes.

Prepare for the 42 School or 1337 Pool with a mentor that explains concepts, guides debugging through questions, and runs mock exams. Its five modes cover onboarding, learning, debugging, code-ownership checks, and exam practice, while refusing to write solutions for you. Using it requires Claude.ai with Custom Skills enabled.

Apply a repeatable engineering cycle to coding tasks regardless of language or technology stack. The skill emphasizes defining the specification before implementation, test-driven development, incremental delivery, and explicit verification before shipping. It also reviews changes across correctness, readability, architecture, security, performance, reliability, and operability to catch scope creep and recurring failure patterns.

The skill first inspects a project's tokens, fonts, components, styling system, and framework, then builds a standalone HTML mockup using that existing visual vocabulary. You can choose low, medium, or high creative scope and review the result in a browser. It will not change application code until you explicitly approve the mockup, and it supports several common frontend stacks without relying on paid design APIs.

Give pr-buddy a pull request number, branch, or URL and it will lead you through five Socratic stations covering the goal, architecture, decisions, risk, and ownership. You answer before the skill reveals its reading, so the review builds on a mental model you can explain. It then hands off to pr-review-toolkit and closes with a three-bullet recap; authenticated gh and the toolkit are required.