An ongoing autonomous portfolio of single-purpose Apify Actors for the Claude Code / AI agent ecosystem. Generators, security auditors, MCP catalogs, feed watchers — all data-only, all public.
Published, runnable, free to try. Pay-per-compute via Apify (most actors complete for cents).
Monitor competitor websites, pricing pages, hiring signals, and social media. Detects changes automatically using SHA256 hashing. Optional AI-powered strategic analysis via Claude. Perfect for SaaS founders, product managers, and market analysts tracking competitive moves.
Track 5 competitors' pricing pages weekly — get a digest in Slack when any plan changes.
Scan any GitHub repo's CLAUDE.md + .claude/hooks/* for hardcoded secrets, prompt-injection patterns, destructive commands, and AI-agent security antipatterns. Free triage, paid full audit.
Audit your CLAUDE.md before open-sourcing the repo — catches hardcoded secrets, dangerous bash, and prompt-injection footguns.
Crawl any documentation site and emit the 2026 GEO-standard llms.txt + llms-full.txt files. Makes your docs machine-friendly for ChatGPT / Claude / Perplexity / Gemini retrieval.
Crawl your docs site once a month and regenerate llms.txt so ChatGPT / Claude / Perplexity index the right pages.
Pick the top open issues from any GitHub repo and generate ready-to-paste Claude Code prompts that walk the model through investigation, fix planning, implementation, and test verification.
Pick the top 5 bugs in a repo, get 5 ready-to-paste Claude Code prompts that walk explore → plan → implement → test.
Scrape top awesome-mcp-server lists, fetch each MCP server's repo stats, and rank by quality (stars / recency / license / activity). Find the right MCP server for your Claude / Cursor / agent project without manually clicking through 500+ READMEs.
Score 500+ MCP servers on quality, then plug only the top 10 into your Claude / Cursor / Cline setup.
Generate a draft CLAUDE.md for any public GitHub repo. Detects stack (JS/Python/Rust/Go), test/build commands, and top-level structure — saves hours of bootstrapping Claude Code in a new codebase.
Fork a new repo? Generate its CLAUDE.md in 30 seconds instead of writing one by hand.
Scaffolded and queued — they ship through the auto-publish pipeline at ~1-3 per day. Email ian.y.mu@gmail.com to fast-track one.
Daily snapshot of Product Hunt #1-10 launches with metadata for trend tracking.
Watch SaaS pricing pages for plan / price / quota changes.
Track LinkedIn company headcount, hiring signals, and post velocity.
Surface trending subreddits and hot threads with engagement velocity scoring.
Aggregate new YC / launch / funding signals into a daily startup radar.
Given a seed GitHub repo, find structurally similar projects by stack and patterns.
Find YouTube videos mentioning Claude Code with view-count and recency scoring.
Mine Hacker News + Reddit threads for explicit pain-point language and frequency.
Fingerprint a GitHub org by AI / LLM / agent dependencies across all repos.
Convert any GitHub repo into a draft Claude Code Skill with proper metadata.
Aggregate the best cursor.directory rules grouped by stack and project type.
Find published Claude Code Skills by capability, language, and quality score.
Detect which AI dev tools a project uses by scanning repo signals.
Curated CLAUDE.md templates for popular stacks (Next.js, Rust, Python, Go).
Daily feed of AI coding / LLM / agent engineering jobs across boards.
Watch the Anthropic blog and emit a structured feed when new posts land.
Monitor the Claude Code changelog and surface breaking / additive changes.
Track pricing changes across Cursor, Windsurf, Copilot, Claude Code, etc.
Discover open-source agent eval harnesses with adoption and recency scoring.
Trend radar for MCP servers — rising stars, falling, stable, with weekly delta.
Find AI podcast episodes by guest, topic, or company mention.
Monitor a Vercel project for build failures and deploy regressions.
Search GitHub Actions templates by language, framework, and use case.
Build an integration graph: which AI dev tools talk to which other tools.
Daily feed of dev.to articles trending for AI / LLM / coding tags.
Track open-source LLM leaderboard rankings (Hugging Face, lmsys, etc).
Track install counts and rankings for VS Code AI extensions.
Watch the public Claude Skills directory for additions, updates, removals.
GitHub trending repositories filtered and ranked by topic / language.
CVE-style security scanner specifically for CLAUDE.md and .claude/ configs.
Track PR velocity / merge times across a repo or org to spot bottlenecks.
Aggregate developer conferences, meetups, hackathons by topic and city.
Curated marketplace catalog of Claude Code hooks with quality scoring.
One feed for all AI dev tool releases (Cursor, Claude, Copilot, OpenAI, etc).
Daily snapshot of careers pages at Anthropic, OpenAI, DeepMind, Mistral, etc.
Historical price/plan timeline for every major AI coding tool.
Track affiliate / referral programs available for AI dev tools.
Find paid GitHub issues across bounty platforms (Algora, Replit, etc).
Every AI agent — Claude Code, Cursor, Windsurf, Cline, custom MCP clients — needs the same boring data: changelogs, pricing pages, repo signals, security audits, server catalogs. Instead of every agent author rebuilding the same scraper for the 1,000th time, I publish them once as single-purpose Apify Actors: callable, versioned, observable, billable per-run.
Each tool does exactly one thing. No framework lock-in, no SDK. JSON in, JSON out. Use them from a curl, a Claude tool call, an n8n workflow, a Zapier zap, or any AI agent that can hit an HTTP endpoint.
The portfolio compounds: when tool #1 ships, it's a curiosity. When tool #100 ships, it's the default data layer for anyone building on top of Claude Code. That's the bet.
I'm bundling the top 20 actors as a single Claude Code Skill Pack — pre-configured, with a unified config schema, a shared cache layer, and one invoice instead of 20. Email me if you'd want first dibs (under $200, beta in May).
Email me to reserveI take custom scraper / catalog / watcher requests. From spec to live Apify Actor in <48h.
Request a custom tool