verbatim-audit-notify
Send Pushover notifications with UUID-linked verbatim JSONL audit trail. TRIGGERS - pushover notify, send pushover, observability alert, verbatim notification, fleet alert, pushover-lookup, audit log notification, push notification with UUID
telemetry-terminology-similarity
Score telemetry field name similarity across syntactic, taxonomic, and semantic layers.
odb-microstructure-forensics
Investigate why ODB bars are oversized, have zero duration, or show anomalous price ranges by forensically analyzing Parquet.
draft-message
Use when an AI agent has drafted a long or sensitive Telegram message and the user wants to review it before sending: posts it to Saved Messages for editing and copy-paste.
dump-channel
user wants to archive, dump, or back up an entire Telegram channel or chat history to NDJSON with all media files downloaded. Full history.
unlimited-ocr-batch-folder
Batch parse a folder of PDFs and images to markdown with LaTeX, tables, and layout boxes using baidu/Unlimited-OCR locally — with checkpoint/resume, incremental writes, and per-file repetition detection. Use when processing a large folder of scanned documents or research papers, when you need results written incrementally to survive interruption, or when extracting equations and tables from a document collection. TRIGGERS - batch ocr, ocr a folder, batch parse documents, parse multiple pdfs, folder to markdown, extract formulas from folder, ocr multiple files, batch document extraction, document collection parsing, checkpoint resume.
unlimited-ocr-parse-document
Parse an image, PDF, or directory into markdown with LaTeX formulas, HTML/markdown tables, and per-block layout bounding boxes, using baidu/Unlimited-OCR locally — free, offline, no API key, no upload quota. Runs on Apple Silicon via MLX (~2.4 s/image, ~5 GB) or an NVIDIA GPU via transformers. Use when transcribing scanned or image-based documents, extracting equations from a paper, converting a PDF to markdown, OCR-ing screenshots, or replacing a paid vision-API OCR call with a local one. TRIGGERS - unlimited-ocr, ocr this, ocr a pdf, extract text from image, transcribe document, pdf to markdown, document parsing, extract formulas, extract equations, latex from image, local ocr, offline ocr, scanned pdf, image to text, layout parsing, bounding boxes from document.
unlimited-ocr-segment-figure
Split a composite figure, screenshot, or dashboard into its individual panels by cropping the layout bounding boxes Unlimited-OCR detects, so each panel can be described separately by a vision model that is good at charts. Turns Unlimited-OCR's refusal to transcribe charts into a pre-processing win — one chart per prompt beats a nine-panel collage per prompt. Use when a vision model is misreading a dense multi-panel image, when you need per-figure crops from a paper or report, or when building a two-stage OCR pipeline that localises first and describes second. TRIGGERS - segment figure, split panels, crop charts from image, multi-panel figure, extract subfigures, per-panel description, layout segmentation, chart crops, split a screenshot into regions, dashboard panels, figure extraction.
azure-provision
>
cf-access-wall
>
dashboard-forge
>
article-writing
Write a complete search-led SEO article from a target keyword. Use when the user wants the finished article, including its preparatory brief and outline, grounded in an existing site's positioning and editorial style.
backlink-outreach
Find and qualify guest-post, niche-edit, or link-exchange opportunities, match them to the user's content, draft tailored outreach, maintain a flexible campaign log, and submit approved requests. Use when the user wants to contact a publisher or page owner for an editorial backlink. Use linking-opportunities for page-only opportunity mapping without outreach.
competitor-intelligence
Analyze SEO competitors and turn SERP, sitemap, and optional provider evidence into strategy. Use when the user wants competitor discovery, a competitor profile, a site comparison, content or keyword gaps, or monitoring over time.
geo-audit-report
Audit GEO and AI-search visibility. Use when the user wants to measure brand mentions, citations, search traces, maps, competitors, or fan-out queries in AI answers and receive a reusable report.
internal-linking
Audit and improve contextual internal linking across a website. Use when the user wants to map links, find orphan or underlinked pages, strengthen a confirmed commercial target or topical cluster, recommend source-to-target links, or apply explicitly approved edits.
linking-opportunities
Find contextual backlink opportunities on a specific prospect site using localized SERPs, page evidence, and outreach angles. Use when the user wants pages on another domain that could link to their content.
posthog-seo-geo-tracking
Set up, audit, and operationalize PostHog tracking from SEO or LLM-referral landing through signup, subscription, and revenue. Use when a user wants SEO/GEO acquisition attribution in a website and SaaS app, a PostHog dashboard for it, or a copy-paste implementation prompt. Do not use for measuring AI-search visibility itself or for query-level SEO reporting without Search Console.
programmatic-seo
Design SEO-driven pages at scale using templates and data. Use when the user wants a programmatic SEO strategy, page-family specification, or implementation for directory, location, comparison, integration, or other template-driven pages.
search-intent-coverage
Build a standalone search-intent analysis or content outline from localized SERP evidence or competitor pages. Use when the user wants to understand and plan coverage without drafting the complete article.
seo-audit-report
Build an interactive Search Console SEO report. Use when the user wants a client-side Vite+React dashboard that imports GSC data, stores it locally, and visualizes query and page KPIs.
seo-context
Create or update reusable SEO context for a site, product, or content program. Use when the user wants to preserve shared SEO facts across tasks, or when another SEO skill needs site, market, priority-page, competitor, or tooling context that has not been recorded yet.
seo-roast
Roast a landing page or article for technical SEO, on-page quality, and search intent. Use when the user wants a direct page critique or an optional screenshot-based HTML report.
subkeyword-injector
Use a page's Search Console queries to propose or apply focused long-tail content edits. Use when the user wants to refresh an existing page based on the queries it already ranks for.
illustration-ideas
Propose specific illustrations that improve comprehension of a web page. Use when the user provides a URL or article and wants visual concepts with placement and layout guidance.
mcp-development
Complete MCP development toolkit for creating, debugging, testing, and reviewing MCP servers. Use when setting up new MCP projects, creating tools, debugging connection issues, reviewing MCP code, or generating documentation.
mcp-expert
Expert in Model Context Protocol (MCP) server development. Use when building MCP servers, creating tools for Claude, implementing resources, debugging MCP connections, or integrating databases with Claude Code.
nodejs-expert
Expert Node.js and TypeScript development assistant. Use when writing, reviewing, or debugging Node.js code, TypeScript projects, async programming, streams, performance optimization, or npm packages.
sqlserver-expert
Expert in Microsoft SQL Server development and administration. Use when writing T-SQL queries, optimizing database performance, designing schemas, configuring SQL Server, or integrating SQL Server with Node.js using mssql package.
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codebase-readiness
This skill should be used to run an Agent-Ready Codebase Assessment — scoring a codebase across 8 dimensions with parallel agents, producing a weighted score (0-100), band rating, and improvement roadmap. Supports Ruby, Python, PHP, TypeScript, JavaScript, Go, Java, Scala, and Rust.
conventional-commits
This skill should be used when creating Git commits to ensure they follow the Conventional Commits specification. It provides guidance on commit message structure, types, scopes, and best practices for writing clear, consistent, and automated-friendly commit messages. Use when committing code changes or reviewing commit history.
doc-audit
Audit codebase documentation for accuracy, completeness, and freshness. Compares docs against actual code structure, auto-fixes small discrepancies, reports structural changes. Works with any language/framework. Companion to agent-ready.
resolve
Review pending drift warnings and fix affected documentation or dismiss false positives
scan
Full-codebase documentation drift scan — find every doc that references code reality incorrectly
setup
Configure doc-sentinel drift detection for a project — detect docs root, watched files, and ignore patterns
gridfinity-baseplate-planner
Use this skill when planning and designing gridfinity baseplates for 3D printing. This includes calculating optimal grid sizes from given measurements, determining how to slice large grids into printable chunks based on printer bed dimensions, and calculating padding requirements for non-exact fits. The skill handles both metric and imperial measurements and provides guidance for using gridfinity.perplexinglabs.com to generate the actual STL files.
process-meeting-transcript
Process raw meeting transcripts from Granola or other sources into structured notes with frontmatter, action items, summary, and formatted transcript. Use this skill when the user asks to process a meeting transcript or provides a raw transcript that needs formatting.
parallel-code-review
This skill should be used when performing comprehensive code reviews using multiple specialized review agents in parallel. It provides patterns for concurrent execution, decision tracking to prevent redundancy, and consolidated reporting. Use when needing thorough review coverage from multiple perspectives (security, architecture, performance) or when reviewing large changesets.
linear-implement
This skill should be used when implementing features from Linear issues with full TDD workflow, automated planning, parallel code reviews (security and Rails best practices), systematic feedback implementation, and automated PR creation with Linear integration. Use when the user provides a Linear issue ID (e.g., "TRA-9", "DEV-123") and wants a complete implementation workflow from issue to PR.
rspec-testing
This skill should be used when writing, reviewing, or improving RSpec tests for Ruby on Rails applications. Use this skill for all testing tasks including model specs, controller specs, system specs, component specs, service specs, and integration tests. The skill provides comprehensive RSpec best practices from Better Specs and thoughtbot guides.
tdd-workflow
Use this skill whenever you are implementing a feature using TDD.
agent-ready
Make a codebase agent-ready by scaffolding AGENTS.md, ARCHITECTURE.md, and docs/ structure, installing regression-aware quality gates, and recommending project-local, stack-specific skills. Analyzes codebase structure, generates documentation artifacts following progressive disclosure patterns, installs report/check/baseline quality-gate commands with merge-base-aware CI and a human-reviewed baseline, and audits existing artifacts for staleness and coherence. Use when improving a codebase for AI agent work.
roier-seo
>
audit-claude-settings
Use when auditing Claude Code settings and env vars against the latest docs and suggest tailored changes
best-practices
Use when about to choose, configure, or refine a tool, library, config format, API pattern, or project setup, or before proposing a design of your own — research current guidance, pitfalls, and prior art first; already knowing an approach, or assuming no prior art exists, is not an exemption. Also use when the user asks about best practices, gotchas, or recommended patterns
blindspot
Use when the user asks for a blindspot pass or to find their unknown unknowns, or signals unfamiliarity with a domain, tool, or codebase area ("never used X", "first time doing Y", "no idea where to start", "don't know what I don't know") before working there. Maps the areas their request leaves unnamed, purpose and shape before mechanics, shows how people usually answer each and why, and asks which to explore next, so unknown unknowns become known unknowns they can prompt with. Recommendations on a tool already chosen belong to best-practices
difference
Compare A vs B, weigh pros and cons of A, B, C, ..., or explain the nuance between sentences — a quick answer from what the model already knows
fuck-over-engineering
Use when the user asks what could be deleted or reduced to simplify a codebase, says "simplify", "over-engineered", or wants a repo-, folder-, or file-wide audit for over-engineering — hunts dead code, reinvented stdlib, needless dependencies, and single-implementation abstractions, reports ranked cuts, applies only the picks. Not a diff review and not a bug hunt
pr
Use when the user explicitly asks to push the current branch and open a PR, rewrite an open PR's body from its commits, or wait for CI and merge it
Page 669 of 1647 · 82344 results