Start with the job, then find an open-source answer

Smart Router is a prompt-only Claude Code skill that classifies work with fast rules and multidimensional scoring, then delegates tasks to Haiku, Sonnet, or Opus through subagents. Multi-step requests can be routed one step at a time, and failed lower-tier attempts can escalate to a stronger model. The goal is to avoid expensive models for simple work without adding a server, package, or runtime service.

ResumeHonest builds an experience library from existing resumes, researches the target role, and can interview the user to uncover relevant work that was never documented. It scores each experience match, pauses for approval at key stages, and produces ATS-oriented Markdown or Word output. Explicit guardrails prohibit fabricated technologies, metrics, titles, or scope, so users need at least one real resume and a job description to begin.

Ask it to draw a call graph, architecture map, dependency view, or workflow for a codebase. The skill analyzes key components, modules, and relationships, then outputs Mermaid diagrams in a consistent `flowchart TD` layout, making it useful for developers who need a quick visual overview of an unfamiliar project.

Add the skill to a Laravel/Statamic project to give your coding assistant conventions for Blade-only templates, the Eloquent Driver’s hybrid storage, content queries, Value objects, navigation, blueprints, and assets. It installs through Laravel Boost or by copying the skill file, and is aimed at teams building or maintaining this specific CMS stack.

Run the CLI against a project to detect its stack, match suitable MCP servers from a curated registry, and check pinned versions, package integrity, install hooks, and multiple advisory feeds. It can also validate stored pins offline. The tool is an install-time gate—not a sandbox, runtime monitor, or guarantee that a third-party server is benign.

Use the skill to examine a consumer iOS or mobile app through frameworks for idea validation, network cold starts, viral loops, onboarding, launch, monetization, and retention. It selects the parts of the playbook that fit the situation instead of loading everything at once, making it useful for founders and product teams shaping a growth strategy.

The skill gathers requirements through layered dialogue about the problem, proposed direction, and measures of success rather than filling fields mechanically. It scores completeness, detects the project type, fills gaps, and then generates a Problem Framing document, SRD, or execution-level PRD with the relevant optional sections. A post-generation review covers product, design, and engineering perspectives.

pr-rules-sync scans merged GitHub pull requests, prioritizes heavily reviewed changes, and weights repeated feedback by reviewer tenure. It turns recurring issues into rules with explanations and real examples, then maintains a marked section in `CLAUDE.md`, `.cursorrules`, `AGENTS.md`, or another instruction file. Re-running replaces only the generated section and preserves handwritten rules; GitHub access and Python 3 are required.

The skill gives you a structured extraction prompt to run in the assistant that already knows you, covering communication preferences, working style, roles, people, workflows, decisions, and active work. Claude then builds a `GLOBAL_CONTEXT.md` file plus role-specific context documents for Projects. It is designed for the Claude desktop app with Cowork, and the resulting Markdown files remain visible, editable, and portable alongside built-in memory.

openclaw-detox checks an agent workspace with health scoring, a disk inventory, rule-compliance audits, and a persistent decisions log. It targets forgotten corrections, false completion reports, wrong-path assumptions, and fragile configuration by requiring work to be saved before it is reported, evidence for completion, snapshots before critical edits, and scripts for rules that keep being violated. It is specifically aimed at improving OpenClaw operations.

Enter a text prompt in this OpenClaw skill to generate an image through the Neta API and receive a single URL for embedding or download. The command supports several size presets and can take a picture UUID as a style reference. It requires Node.js 18 or later and a valid NETA_TOKEN.

Turn scattered ideas into a Notion workflow that moves each item through Spark, Concept, Plan, Active, or Frozen. Claude handles quick capture, triage, a weekly review ritual, and quarterly reflection while enforcing a limit of three active ideas. Setup requires a compatible Notion database, its database and data-source identifiers, and a connected Notion MCP server.

Import a PDF by rendering every page as a JPEG and inserting the images as full-width blocks in a new Notion page, creating a presentation-like scrolling view. The command can set the title, reading status, and tags and can add generated cover and icon artwork. A Notion token and database ID are required; an Unsplash key is optional for cover images.

Keep frontend changes aligned with Feature-Sliced Design by checking layer hierarchy, downward-only imports, slice-level public APIs, segment folders, and naming conventions. The skill can also scaffold a component or feature slice with the expected structure. It supports React Router, Next.js, Remix, and Vite React, with partial coverage for Astro's pages layer.

Use public REST endpoints without signing up or supplying an API key for workflow tasks such as human approval forms, webhook capture, and temporary storage, plus utilities for hashing, encoding, UUIDs, time, and IP lookup. The repository includes Python and JavaScript clients, function-calling definitions, and a Claude skill. Temporary storage, short links, and several webhook records expire automatically after 24 hours.

Start with the lightest launch option for a local HTML folder or development server: create a temporary Cloudflare Tunnel URL, move to a persistent Pages deployment when needed, and optionally connect a domain the user purchases separately. Tunnel links stop working with the process. Pages requires a Cloudflare account and API token, while domain purchase remains manual and automated DNS setup needs a separate scoped token.

PixelPitch converts a LinkedIn visual brief into a prompt package tailored to the selected image model. It can include a main prompt, negative prompt where supported, dimensions and settings, a brand-alignment check, two creative variants, and accompanying post copy. Marketers can also provide brand colors, style preferences, audience details, and visual exclusions for more consistent output.

doc-generator lets Claude collect the required fields, confirm them, and pass a JSON payload to a local deterministic renderer. It currently supports purchase orders, invoices, requests for quotation, and test request forms, with validation and computed totals handled outside the model. Local Python setup is required, and WeasyPrint may need Pango system libraries on macOS or Linux.