Agent skills for StackOne — integration infrastructure for AI agents. 10,000+ production-ready actions, 200+ connectors, and an AI integration builder to extend to any API.
# Add the marketplace
/plugin marketplace add stackonehq/agent-plugins-marketplace
# Install the StackOne plugin (all 7 skills)
/plugin install stackone@stackone-marketplace# Install all skills (works with Claude Code, Cursor, Codex, Windsurf, etc.)
npx skills add stackonehq/agent-plugins-marketplace
# Install a specific skill
npx skills add stackonehq/agent-plugins-marketplace@stackone-agents| Skill | What it does | When to use |
|---|---|---|
stackone-platform |
Platform operations — API keys, accounts, logs, debugging | "Set up StackOne", "list my accounts", "debug API errors" |
stackone-connect |
Account linking via Connect Sessions and the Hub component | "Connect a provider", "embed the integration picker" |
stackone-agents |
Build AI agents with TypeScript/Python SDK, MCP, or A2A | "Add StackOne tools to my agent", "set up MCP" |
stackone-cli |
Custom connector development and deployment | "Build a custom connector", "deploy my connector" |
stackone-connectors |
Discover connectors, actions, and integration capabilities | "Which providers does StackOne support?" |
stackone-unified-connectors |
Build unified connectors that transform provider data into standardized schemas | "start unified build for [provider]", "map fields to schema" |
stackone-defender |
Detect prompt injection and jailbreak attacks in text using local ML | "scan for prompt injection", "is this text safe?", "protect my agent" |
Each skill includes step-by-step workflows, concrete examples, and troubleshooting for common errors.
These skills teach workflows while pointing to live documentation for details that change frequently:
- Step-by-step instructions for common tasks (not just API reference)
- Real user scenario examples with trigger → actions → result
- Error handling and troubleshooting for common failure modes
references/directories for detailed lookup tables loaded on demand- Agents fetch the latest docs, SDK READMEs, and API specs at runtime
- The canonical source of truth remains docs.stackone.com
StackOne publishes a machine-readable documentation index at docs.stackone.com/llms.txt — agents can fetch this to discover all available documentation pages.
- StackOne Documentation
- StackOne Dashboard
- AI Playground
- TypeScript SDK (
@stackone/ai) - Python SDK (
stackone-ai) - CLI (
@stackone/cli) - Hub React Component (
@stackone/hub)
MIT