This is infrastructure for agents that need to spend money and interact with the world as autonomous entities. It gives your agent a wallet (fund via Stripe or USDC), an email address, a phone number, and pay-per-use access to web search, image/video/music generation, X/Twitter search, SMS, voice calls, and 100+ LLM models. The security section is unusually detailed and worth reading: it explicitly calls out prompt injection risks in search results and inbound messages, credential exfiltration via email/SMS, and financial guardrails. If you're building an agent that needs to do more than read-only API calls, this solves the "how does it pay for things" problem, but you'll want human approval hooks around anything that spends money or sends messages.
npx -y skills add atxp-dev/cli --skill atxp --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
sentry/dev
degausai/wonda
github/awesome-copilot
insforge/agent-skills
github/awesome-copilot
breaking-brake/cc-wf-studio