메뉴
Social media account setup guide for AI/developer platforms. Agent prepares all content (bio, avatar, posts) for user to copy-paste, since automation is blocked by Cloudflare/CAPTCHA/React shadow DOM. Supports X/Twitter, Medium, Dev.to, Reddit, LinkedIn.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
Fix missing .dylib libraries when brew upgrades break Node.js browser tooling — create symlinks from old version numbers to new.
Auto-detect environment and record terminal/GUI demos for GitHub repos. Supports VHS, asciinema+agg, svg-term, and cua-driver modes. Single-command orchestration layer.
Use when releasing a new open-source project or skill pack on GitHub. Follows the proven exposure playbook from Hermes Core Skills release — covering README, community health, topics, social media, and verification. Ensures every release meets a minimum exposure bar so the project is discoverable, trustworthy, and ready for adoption.
Multi-angle GitHub project discovery using browser + GitHub API + trending, with practical clone/install/absorption workflow
Use when installing any new tool/package from GitHub, Brew, or Cask — includes post-install verification script, automated health check cron, and deliverable report so the user can independently confirm the tool is in active daily use.
Complete iOS App from Source Code to Build fully automated pipeline. Uses XcodeGen + xcodebuild CLI, no Xcode GUI required.
SOC 직업 분류 기준
| name | social-media-account-registration |
| description | Social media account setup guide for AI/developer platforms. Agent prepares all content (bio, avatar, posts) for user to copy-paste, since automation is blocked by Cloudflare/CAPTCHA/React shadow DOM. Supports X/Twitter, Medium, Dev.to, Reddit, LinkedIn. |
| author | Hermes Agent |
| license | MIT |
| metadata | {"hermes":{"tags":["social-media","registration","signup","accounts","content-planning"],"related_skills":["github-demo-recording","reddit-post-launch-workflow"],"tokens":{"scan":40,"load":180,"category":"tool"}}} |
Agent prepares all content for social media account setup. User does the actual form filling because automation is universally blocked.
Most platforms block browser automation, but NOT all. Hacker News can be fully automated with Cua Driver.
Known exceptions that WORK with Cua:
| Platform | What works | Caveat |
|---|---|---|
| Hacker News | ✅ Full registration + Show HN submission | No CAPTCHA, no email, just username+password. Title limited to 80 chars. |
| Mastodon (generic) | ✅ Form filling | Needs email verification link click (human step) |
Known blocks (the rest): This was learned through ~30 minutes of failed attempts across X, Medium, GitHub, and Reddit — every single one blocked.
Additional hard-earned lesson: user's bio needs to be future-proof. When writing bios for GitHub/Medium, don't tie the bio to a single project (e.g. "Creator of Hermes Core Skills"). Use a broader, evergreen description like "Building useful tools for AI agents and developers. Open source." that won't become outdated when new projects launch. Each platform may need a slightly different version — X needs specificity (what are you doing NOW), GitHub needs generality (who ARE you), Medium needs writing-related angle.
| Platform | Block Type | What happens |
|---|---|---|
| X/Twitter | React shadow DOM + bot detection | Programmatic clicks ignored, blank screen after Google OAuth |
| Medium | Cloudflare | "Performing security verification" — page never loads |
| Cloudflare + CAPTCHA | "You've been blocked by network security" | |
| Dev.to | Cloudflare | Same as Medium |
All platforms also require SMS/email verification codes or CAPTCHA — human-only steps that cannot be bypassed even with semi-automation.
Agent prepares: User does:
┌──────────────────────┐ ┌──────────────────┐
│ • Pre-written bio │ ──→ │ • Go to signup URL │
│ • Avatar file ready │ ──→ │ • Upload avatar │
│ • First post content │ ──→ │ • Copy-paste bio │
│ • Thread templates │ ──→ │ • Copy-paste posts │
│ • Content calendar │ ──→ │ • Solve CAPTCHA │
│ • Unified identity │ ──→ │ • SMS verify │
└──────────────────────┘ │ • Enter password │
└──────────────────┘
x.com/signupx.com/settings/profilemedium.com → "Sign in with Google" (best path)medium.com/me/settingsmedium.com/new-story or from profile menu → "Write a story"dev.to/enter → "Sign in with GitHub"dev.to/settingsdev.to/newreddit.com/registerlinkedin.com/signup| Platform | Bio | Why |
|---|---|---|
| GitHub | Building useful tools for AI agents and developers. Open source. | Evergreen, not tied to any single project |
| X/Twitter | Building tools for AI agents. Creator of Hermes Core Skills. Open source. | Needs specificity — people scan and need to know your thing |
| Medium | Building tools for AI agents & developers. Writing about open source, AI agents, and developer workflows. | Writing platform, signals you'll produce content |
| Dev.to | AI agent tooling & open source. Building Hermes Core Skills. | Developer community, direct is fine |
| (leave blank) | Reddit doesn't prioritize bios | |
| Building AI agent tools at Nous Research. Creator of Hermes Core Skills. Open source contributor. | Professional tone, full context |
⚠️ Never lock your bio to a single project. User specifically corrected this: "如果我以後再發佈新 project 咪唔相符". Use general descriptions like "building tools for AI agents" that scale across multiple projects.
| Field | Value |
|---|---|
| Display Name | Chris Lam |
| Username | ChrisLamDev |
| Avatar | Geometric CL monogram (orange-red C + green-teal L on vibrant gradient bg) |
| Bio (full) | Building AI agent tools. Creator of Hermes Core Skills — 25 executable skills for debugging, planning & token efficiency. |
| Bio (short, X) | Building AI agent tools. Creator of Hermes Core Skills — 25 executable skills for debugging, planning & token efficiency. Open source. |
Just dropped Hermes Core Skills — 25 executable AI agent skills for debugging, planning, token efficiency & security.
🚀 `cp -r skills/* ~/.claude/skills/`
⭐ github.com/ChrisLamDev/hermes-core-skills
Built for Claude Code, Codex, Cursor & Hermes. #AIAgents #OpenSource
1/N Let's talk about AI agent reliability.
Most people think better prompts = better outputs.
But there's a better way: skills.
2/N A prompt is a one-shot instruction. A skill is a repeatable workflow with:
• Triggers (when to use it)
• Steps (how to execute)
• Checks (how to verify)
3/N Hermes Core Skills turns this into executable files:
skills/debugging/SKILL.md
skills/planning/SKILL.md
skills/review/SKILL.md
4/N One command to install all 25:
cp -r skills/* ~/.claude/skills/
Open source, MIT licensed.
github.com/ChrisLamDev/hermes-core-skills
Use Python PIL to generate a CL monogram avatar. Expect 5-7 iterations — user feedback loop is essential:
| Iteration | Feedback | Fix |
|---|---|---|
| v1-v2 | "C and L not visible" | Arc/path rendering issues, switch to solid blocks |
| v3 | "Too dark/somber" | Black bg → vibrant gradient (red-orange-purple) |
| v4 | "Not sharp enough" | Boost saturation, add high-contrast accent dots |
| v5 | "Not harmonious" | Reduce saturation, blend colors smoothly |
| v6 | "No tech feel" | Add grid overlay, outer glow ring, neon-style accents |
| v7 | ✅ "Good" | Solid block letters (orange C + green L), vibrant bg, subtle glow |
Final approach that worked:
| Day | Platform | Content Type |
|---|---|---|
| Day 1 | X + Medium | Launch post + full article |
| Day 2 | X | Value post (token efficiency) |
| Day 3 | X | Tutorial (debugging workflow) |
| Day 4 | Community post (what I learned) | |
| Day 5 | X | Show & tell (most popular skill) |
| Day 6 | X | GitHub star update |
| Day 7 | X | Weekly roundup |
x.com/compose/post — 404 if not logged in, user must be authenticated first.medium.com/new may 404 — use "Write a story" from profile menu instead.reddit.com first then clicking "Sign Up" can work.