Edit images with PIL/Pillow.

Tiered safety scanner for URLs AND email attachments.

Audits `.mcp.json` and Claude Code MCP config files across a workspace for plaintext secrets, hardcoded API keys (instead of `${VAR}` interpolation), suspicious npx…

Scans web page content, document text, OCR output, or any untrusted text for prompt-injection patterns (hidden instructions, role-override attempts, "ignore previous"…

Git-based snapshot of your ~/.claude directory (skills, settings, memory, singularity scores).

Diagnoses every skill in ~/.claude/skills/ for common problems and offers safe auto-fixes.

Fractional business coach in a skill.

Generates filtered business idea lists from inputs (skills, capital, time, lifestyle constraints).

Acts as a fractional CISO (Chief Information Security Officer) for SMB and mid-market organizations.

Acts as a fractional CTO for SMB and mid-market organizations.

Research-and-recommend skill for PCs, laptops, monitors, keyboards/mice, headsets, network gear (routers/switches/APs/firewalls), NAS, printers/MFPs, UPS, and IT…

Fractional marketing coach in a skill.

Meeting preparation coach in a skill.

Reviews social-media post drafts and account behavior against the published policies of Meta (Facebook + Instagram), TikTok, YouTube, X / Twitter, LinkedIn, Threads,…

Three-in-one risk skill - **Analyst** identifies risks, **Assessor** rates likelihood x impact, **Mitigator** proposes controls.

Fractional sales coach in a skill.

Recommends and compares business software stacks (CRM, helpdesk, accounting, project mgmt, password manager, RMM, BCDR, email, collaboration, design, payroll, AI tools).

Customer support response coach in a skill.

Long-form writing companion for fiction novels, nonfiction books, online courses, and serialized stories.

Generates LinkedIn carousels for any brand.

Edit audio files with ffmpeg.

Edit videos with ffmpeg.

Universal "make this file smaller" tool.

One skill, four modes for working with a video.

Extracts key scene screenshots from YouTube videos at major topic transition points.

Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute.

Write JavaScript code in n8n Code nodes.

Write Python code in n8n Code nodes.

Validate n8n expression syntax and fix common errors.

Set up Husky pre-commit hooks with lint-staged (Prettier), type checking, and tests in the current repo.

"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files).

Creates Excel (.xlsx) workbooks for any purpose using openpyxl.

Debugs broken Excel formulas.

Takes a folder of mixed file types (PDFs, DOCXs, images, emails, screenshots, audio, video) and auto-routes each to the appropriate skill (pdf-redactor, file-compressor,…

Helper for Jupyter notebooks (.ipynb) and structured note-taking.

Permanently redacts sensitive data from PDFs (PII auto-detect or manual regions), then rasterizes each page to an image-only PDF so the redaction CANNOT be undone,…

Convert any HTML file to a print-ready PDF (brand colors preserved via Playwright Chromium) and/or a structured DOCX with proper page breaks, heading styles, tables, and callout boxes.

One-stop PDF manipulation toolkit using pypdf (pure-Python, no LaTeX).

Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui).

"Use this skill any time a spreadsheet file is the primary input or output.

Generates a comprehensive Word document training guide from any online course URL.

Encouraging general-wellness coach for physical habits (sleep, hydration, movement, nutrition basics, screen breaks, ergonomics).

Aggregates job listings from multiple FREE sources (USAJobs.gov, RemoteOK RSS, We Work Remotely RSS, Hacker News "Who is Hiring" parse, Adzuna API if key set,…

Supportive, encouraging skill for everyday mental wellness, gentle motivation, guided meditation scripts, gratitude practice, breathing exercises, and study/stay-strong…

Reflective personality coach grounded in research-backed assessments.

Generate quiz items (multiple-choice, true/false, short-answer, scenario / case-based, fill-in-blank, matching, drag-and-drop, drag-image) for training, certification…

Updates an EXISTING resume DOCX in place - adds a new role/position, refreshes dates on current role, swaps in new skills, updates contact info, or appends a new…

Create exercise directory structures with sections, problems, solutions, and explainers that pass linting.

Socratic study companion for high-school and college students.

Teaching assistant for K-12 and higher-ed instructors.

Audits your calendar density, finds focus blocks, drafts time-blocking proposals.

Blocks password entry on look-alike auth domains.

Reviews installed Chrome, Edge, and Firefox extensions for risky permissions.

Scans the system clipboard for secrets, PII, or credentials BEFORE you paste anything sensitive into an AI chat (Claude, ChatGPT, Gemini), public form, screenshot, or…

Scans `package.json`/`package-lock.json`, `requirements.txt`/`Pipfile.lock`, `go.mod`/`go.sum`, `Gemfile.lock`, `Cargo.lock`, and `pyproject.toml` against OSV.dev (free,…

Audits a sending domain's email-authentication DNS records — SPF, DKIM (probes common selectors), DMARC, MTA-STS, BIMI, and DNSSEC.

Scans file attachments for malware.

Tiered URL safety scanner.

Inspects HTML forms for fields hidden via type=hidden, display:none, visibility:hidden, off-screen positioning, zero-size, or aria-hidden, then flags any hidden field…

Scans LLM-generated output (code, suggestions, docs) for references to files, functions, classes, or packages that do NOT actually exist in the project.

Tiered URL safety scanner for ANY source — emails, web research, pasted links, social DMs, chat messages, search results.

Detects bulk/mass-send patterns before any outbound message.

Audits OAuth consent screens for over-broad scope requests.

Single-call safety check before any outbound message — email, Slack, LinkedIn DM, SMS, Teams, public post.

Diagnoses what happened to an email by parsing its raw message headers.

Checks if a password (or list of passwords) appears in known data breaches via the Have I Been Pwned (HIBP) Pwned Passwords API.

Redacts personally identifiable information (PII) from text — emails, phone numbers, US SSNs, credit cards (Luhn-validated), street addresses, IP addresses, dates of…

Warns when clicking or rendering a link will leak the current URL via the HTTP Referer header.

Scans files, folders, git diffs, clipboard text, or any string for leaked credentials — API keys, OAuth tokens, AWS keys, GitHub tokens, Stripe keys, JWT, private keys,…

Analyzes a message (email body, SMS, Slack DM, voicemail transcript, LinkedIn message) for social-engineering and manipulation patterns - urgency tactics ("respond…

Follows HTTP redirects from a shortened URL (bit.ly, tinyurl, t.co, etc.) up to N hops without executing JavaScript, showing every hop in the chain and the final…

One-command installer for popular community + official skill marketplaces.

Guides you through obtaining an API key for any skill that needs one, validates the key works, and stores it securely in the OS keychain (Windows Credential Manager,…

The "set up Claude right the first time" wizard.

The 60-second sprint.

Viewer + analyzer for the Claude Code hook event log at ~/.claude/hook-log.jsonl.

Fresh-machine installer for HumanFirst.

Reads + verifies dependency declarations across all installed skills.

Renders every markdown doc in the HumanFirst pack (or any folder) into a self-contained, browsable HTML site with a sidebar tree, search, and clean print-friendly…

Validates skills + packs against the four-layer security contract from ARCHITECTURE.md.

Weekly drift + reliability check for your HumanFirst setup.

The weekly/monthly batch improvement loop.

Viewer + analyzer for the intent-guard audit log at ~/.claude/intent-guard.jsonl.

Curated catalog of vetted Claude Code skill marketplaces (Anthropic official, Cowork knowledge-work plugins, plus 9 vetted community marketplaces with 600+ skills total).

Searches your installed skills by intent ("I want to write a proposal", "I need to scan an email") and returns ranked matches with the exact CLI invocation.

Turns a natural-language goal ("onboard a new client", "audit client website monthly", "publish a LinkedIn post end-to-end") into a planned workflow DAG that composes…