CLI + Skills

Set up AI agents to build and manage superglue tools via the CLI and superglue skill

AI coding agents work best when they understand your systems, APIs, and tools. The superglue CLI ships with a skill — a structured reference that teaches agents how to use every sg command, handle auth, build tools, and debug failures. Install once, works across 35+ agents.

---

Installing the CLI

npm install -g @superglue/cli

Setup

Interactive setup:

sg init

Prompts for your API key (get one at app.superglue.cloud/admin), endpoint, and output preferences.

Environment variables (CI, AI agents, non-interactive):

export SUPERGLUE_API_KEY="your-api-key"
export SUPERGLUE_API_ENDPOINT="https://api.superglue.cloud"  # optional, this is the default

Verify: sg system list — if you see fetch failed or auth errors, double-check your API key and endpoint.

---

Installing the Skill

The skill gives your AI agent full CLI knowledge — commands, auth patterns, tool schemas, debugging, and deployment. Install once and the agent references it automatically.

Universal Install (all agents)

npx skills add superglue-ai/cli

Auto-detects your installed agents. Target a specific agent with -g -a <agent>:

npx skills add superglue-ai/cli -g -a claude-code
npx skills add superglue-ai/cli -g -a codex
npx skills add superglue-ai/cli -g -a cursor

Agent-Specific Install

Claude Code

From the Anthropic marketplace:

/plugin install superglue@claude-plugins-official

Or load from local install:

claude --plugin-dir $(npm root -g)/@superglue/cli

Codex

Via the skill installer:

$skill-installer --repo superglue-ai/cli --path skills/superglue

Or manually:

cp -r $(npm root -g)/@superglue/cli/skills/superglue ~/.agents/skills/superglue

Check it into your repo at .agents/skills/superglue/ so the whole team gets it.

Cursor

cp -r $(npm root -g)/@superglue/cli/skills/superglue .cursor/skills/superglue

Or global: cp -r $(npm root -g)/@superglue/cli/skills/superglue ~/.cursor/skills/superglue

---

What the Skill Provides

• All CLI commands with flags, options, and usage patterns
• Authentication patterns — credential placeholders, OAuth flows, header formats
• Tool configuration schema — step configs, data selectors, transforms, pagination
• Debugging workflows — common errors, --include-step-results, sg system call for isolation
• Deployment patterns — SDK, REST API, webhooks (via references/integration.md)
• Specialized references for databases, file servers, and transforms (loaded on demand)

---

Adding Project-Specific Context

The skill teaches the AI how to use superglue. Add a section to your project’s AI config file (CLAUDE.md, AGENTS.md, or .cursorrules) so it knows to use the skill and discover your setup dynamically:

## Using Superglue

When working with superglue tools — via the CLI, SDK, or REST API:

1. Invoke the superglue skill and read the SKILL.md file before running any sg command
2. Read the relevant reference files for the task (e.g. http, postgres, file-handling, integration)
3. Ensure the CLI is configured (sg init has been run or env vars are set)
4. Run sg system list and sg tool list to discover the current setup before building or modifying anything
5. Never hardcode system IDs or tool IDs — always discover them dynamically

---

Best Practices

Always test systems before building tools

Have the agent run sg system call to verify auth and see the actual response shape before constructing a tool config:

sg system call --system-id stripe \
  --url "https://api.stripe.com/v1/customers?limit=2" \
  --headers '{"Authorization":"Bearer <<stripe_api_key>>"}'

Use --include-step-results when debugging

Shows the raw API response from each step — pinpoints whether the issue is auth, the endpoint, or a transform:

sg tool run --draft <id> --payload '{}' --include-step-results

Never paste secrets into chat

Use --sensitive-credentials when creating systems via the CLI so secrets are prompted securely:

sg system create --id my_api --url https://api.example.com \
  --credentials '{"client_id":"abc"}' \
  --sensitive-credentials client_secret,api_key

---

LLM-Readable Documentation

For agents that don’t support skills, we provide full docs in LLM-friendly formats:

File	URL	Purpose
Index	llms.txt	Lightweight index of all pages
Full docs	llms-full.txt	Complete documentation in one file

If your tool can fetch URLs — give it the URL. If it requires file uploads — download llms-full.txt and upload directly.

---

Support & Resources

• skills.sh — Install the superglue skill for any AI agent
• CLI Reference — Full command documentation
• Discord Community — Get help from other developers
• GitHub Issues — Report bugs or request features