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

AuditKit fetches an authorized site’s robots.txt, sitemap, HTML, headers, and JSON-LD, then uses deterministic checks for SEO, performance signals, accessibility, conversion, security headers, structured data, and AI-search readiness. It applies a published weighted rubric and returns prioritized, copy-ready fixes, while clearly flagging limits such as backlinks, rankings, field performance data, authenticated pages, and content the environment cannot render.

Public Speaking Coach starts with the audience, occasion, intention, supporting material, and delivery constraints. It selects an occasion-specific structure, shapes an engaging opening and concise body, and adds annotations for hooks, pauses, stories, appeals, callbacks, and the intended audience effect.

ai-usage-mirror reads local Claude Code and Codex session files, normalizes them into SQLite, and summarizes models, projects, commands, correction patterns, and recurring tasks. It is local and read-only, producing a retrospective report page rather than a real-time dashboard or a reader for server-side desktop chats.

ga-cli exposes Google Analytics 4 through task-first commands that return flattened JSON and stable exit codes. Its read-only surface covers reports, realtime activity, pages, sources, events, trends, property metadata, compatibility checks, quota checks, and credential diagnostics without interactive prompts.

This skill builds whiteboard explainers in HyperFrames with isolated artwork, serpentine mask reveals, and a photographed hand holding the marker. Whisper word timestamps drive drawings, labels, camera moves, and sound cues, while coverage checks, timed frame audits, and narration verification catch visible failures. It also requires Python, Whisper, Pillow, an image model, and a TTS voice.

safe-migrations reviews PostgreSQL and MySQL schema changes for locks, table scans, and downtime risks before they merge. It explains the blocking mechanism, rewrites risky operations into copy-ready staged steps in SQL or framework idioms, and sequences expand-contract deployments when application changes must remain compatible.

This project pairs an MCP readiness audit with a server blueprint. The audit checks whether analytics and funnel data is reachable through permitted tools, while the blueprint guides a read-only interface with access controls and privacy boundaries; the guide also explains what MCP can and cannot change about product outcomes.

Life Design Skill uses a Socratic sequence of six to nine questions to separate fixed constraints from problems that can be shaped. It produces three distinct five-year Odyssey Plans, a personal blueprint grounded in the user’s own words, and an offline HTML dashboard for the resulting reflections.

bili-download archives individual Bilibili videos or full creator collections into ordered, season-based folders. It supports QR-code login, incremental updates, command-line use, and agent invocation; its session cookie file must remain private and excluded from version control.

IconLoom scans project icons into an editable SVG workbench for Claude or Codex. You can change vector details manually or ask the model to redraw them to a shared specification, reuse assets across projects, recognize system symbols, and write supported changes back to source files across iOS, Android, web, mini-program, and React Native projects.

This Claude skill prepares a Vietnam software copyright dossier for web or mobile projects. It scopes pages, owners, and authors, captures interfaces with Playwright, redacts sensitive data and third-party marks, renders source-code images, and assembles Markdown, DOCX, and paginated PDF files. It is a workflow aid rather than legal advice, so current filing requirements must still be checked with Vietnam’s copyright authorities.

substack-migration-check inspects a Substack export ZIP without modifying the original. It creates an integrity and dependency report, cleaned and excluded subscriber CSV files, and a post inventory for hosted images, links, embeds, and paywalls. Processing stays offline: a local script handles subscriber data, while the model sees only aggregate counts and redacted samples.

This skill guides accessible charts in R and Python with color-vision-safe palettes, redundant shapes or line styles, direct labels, contrast checks, and simulations for common color vision deficiencies. Its references cover popular plotting libraries, while a script checks WCAG non-text contrast and CIELAB color distance.

SlopCheck flags recurring AI-style habits such as fake profundity, self-narration, forced reframes, and mic-drop endings. Its Guide, Detect, and Rewrite modes can prevent those patterns, score a draft, or return plainer English with a brief note on what changed; portable rules work in other tools too.

tableau-semantics-claudecode-skill packages Tableau Semantics Marketplace workflows as a shareable Python CLI. It runs health checks and produces performance, fleet, revenue, or utilization reports in Markdown or JSON from mock, live Tableau, or hybrid sources. Mock mode works without a live connection; real Tableau access requires administrator configuration, and failed requests intentionally fall back to preserve the workflow.

journal-fit-finder extracts a manuscript’s field, methods, article type, readership, language, and constraints, then ranks journal candidates with a transparent fit rubric. It checks scope, recent precedent, indexing, APC, publisher diversity, and rejection risks, while refusing to invent acceptance rates or quartiles. Journal policies, fees, rankings, and turnaround times change, so final targets still need current primary-source verification.

cold-email-doctor reads DNS, IMAP, or supported sending-platform data to check volume, bounces, warmup, SPF/DKIM/DMARC, shared IPs, and inbox placement. Its connectors are read-only and turn the findings into root-cause fixes. DNS checks need no API key, while platform checks require the relevant credentials; the tool cannot change settings or send email.

Claude Session Miner reads local Claude Code JSONL transcripts, strips tool-call noise, and extracts durable ideas, decisions, pending items, and system facts into one JSON object per line. A per-host registry prevents duplicate processing, so it can run through launchd, cron, or systemd. Apart from its inbox and registry it only reads transcripts, and its extraction prompt explicitly excludes secrets, tokens, and personal data.