agentleFS
intercomintercom/claude-plugin-external/skills/install-cli/SKILL.md

install-cli

Skill3 starsChanged 8 months agoMITView on GitHub

Install and authenticate the Intercom CLI (`@intercom/cli`) — a command-line tool for managing Intercom workspaces, designed for both human operators and AI agents. Use when the user asks to "install the Intercom CLI", "set up the intercom command", "install @intercom/cli", or wants shell access to their Intercom workspace.

  • Installs packages

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before installing, verify the user has:

    line 19
  2. Very specific:

    If node --version returns < 20.6.0, stop and ask the user to upgrade before proceeding.

    line 24
  3. Very specific:

    If the user gets a permission error (EACCES) on macOS or Linux, they likely have a system Node install.

    line 38
  4. Very specific:

    Do not suggest sudo npm install -g — installing global packages as root is a known footgun and will cause permissions issues on later …

    line 42
  5. Very specific:

    Add this to ~/.zshrc, ~/.bashrc, or the equivalent so it persists across shells.

    line 71
  6. Distinctive:

    If the user has additional setup needs (importing articles, enabling Fin, etc.), point them at:

    line 85
  7. Distinctive:

    After install, the user can run any of these.

    line 126
  8. Very specific:

    If the user is using the CLI from a Claude Code agent or other automation, prefer --json or --jq over parsing human-formatted output.

    line 156
  9. Distinctive:

    If the user manages multiple workspaces:

    line 160

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Install Intercom CLI

Install the @intercom/cli package globally and authenticate it against the user's workspace. The CLI complements this plugin: the plugin is for in-conversation MCP usage, while the CLI is for shell access, scripting, and workspace provisioning.

Prerequisites

Before installing, verify the user has:

  1. Node.js >= 20.6.0 — Run node --version to check. If missing or older, direct them to install via nodejs.org or their version manager (nvm, fnm, volta).
  2. npm — Comes bundled with Node.

If node --version returns < 20.6.0, stop and ask the user to upgrade before proceeding. The CLI will not run on older versions.

Step 1: Install Globally

npm install -g @intercom/cli

Verify the install:

intercom --version

If the user gets a permission error (EACCES) on macOS or Linux, they likely have a system Node install. Recommend either:

  • Switching to a user-managed Node via nvm/fnm/volta (preferred), or
  • Configuring an npm prefix under their home directory.

Do not suggest sudo npm install -g — installing global packages as root is a known footgun and will cause permissions issues on later updates.

Step 2: Authenticate

There are two paths. Ask the user which applies.

Path A: Existing Workspace (most common)

The user has an Intercom workspace and needs to connect the CLI to it.

  1. Direct them to the Developer Hub.
  2. Click New app (or select an existing app).
  3. Under Authentication, copy the Access Token.
  4. Authenticate one of two ways:

Option 1 — Persistent (recommended for interactive use):

intercom auth login --token "<paste-token-here>"

This stores the token in the OS keyring (Keychain on macOS, Secret Service on Linux, Credential Manager on Windows) so future commands don't need the env var.

Option 2 — Environment variable (recommended for CI / scripting):

export INTERCOM_TOKEN="<paste-token-here>"

Add this to ~/.zshrc, ~/.bashrc, or the equivalent so it persists across shells. For CI, set it as a secret in the pipeline config.

The env var takes precedence over the keyring-stored credential when both are present.

Path B: New Workspace (provisioning)

The user wants to create a brand-new Intercom workspace from scratch — no token needed.

intercom setup --company-name "Acme"

This command provisions a workspace, signs the user in, and stores the resulting credentials. The CLI will prompt for an email, password, and verification code during the flow.

If the user has additional setup needs (importing articles, enabling Fin, etc.), point them at:

intercom setup --help

Step 3: Verify

Run:

intercom me

Expected output: the admin's name, email, and the active workspace ID. If this fails:

ErrorLikely CauseFix
Not authenticatedNo token in keyring or envRe-run Step 2
401 UnauthorizedToken revoked or wrong workspaceGenerate a new token in the Developer Hub
command not found: intercomGlobal install path not on $PATHRun npm config get prefix, ensure <prefix>/bin is on $PATH

Step 4 (Optional): Shell Completions

Recommend this only if the user uses intercom interactively in their shell.

# zsh — append to ~/.zshrc
intercom completion zsh >> ~/.zshrc

# bash — append to ~/.bashrc
intercom completion bash >> ~/.bashrc

# fish
intercom completion fish > ~/.config/fish/completions/intercom.fish

Restart the shell or source the rc file for completions to take effect.

Quick Reference

After install, the user can run any of these. Don't dump this whole list at them — surface relevant commands based on their goal.

CommandPurpose
intercom meShow current admin + workspace
intercom api <endpoint>Raw API access (like gh api)
intercom articles list|get|searchHelp center articles
intercom contacts list|get|searchContacts
intercom conversations list|get|searchConversations
intercom fin manifest|enable|downloadFin AI Agent
intercom messenger get|update|snippetMessenger config
intercom config listShow CLI config

Full reference: @intercom/cli README.

Agent / Pipe Mode

The CLI auto-detects when stdout is piped and switches to compact NDJSON (one JSON object per line). This makes it ergonomic for scripts and AI agents:

# Auto NDJSON when piped
intercom articles list | jq '.title'

# Force JSON
intercom articles list --json

# Inline jq filter
intercom articles list --jq '.data[].title'

If the user is using the CLI from a Claude Code agent or other automation, prefer --json or --jq over parsing human-formatted output.

Multi-Workspace

If the user manages multiple workspaces:

intercom auth list                  # show stored credentials
intercom auth switch <workspace>    # change default
intercom me --workspace <id>        # one-off override

Each workspace's credential is stored separately in the keyring.

Troubleshooting

npm install -g Hangs or Fails on Corporate Network

The user's npm registry may be proxied. Check npm config get registry — if it points at an internal mirror, that mirror may not have @intercom/cli published. Set the registry explicitly for this install:

npm install -g @intercom/cli --registry=https://registry.npmjs.org

intercom Command Found But Crashes Immediately

Usually a Node version mismatch. Run node --version and confirm it's >= 20.6.0. If using nvm, ensure the active version (nvm current) is the one the global install used.

Token Stored But Commands Still Fail With 401

The INTERCOM_TOKEN env var overrides the keyring. Run unset INTERCOM_TOKEN and try again — a stale env var may be shadowing a fresh keyring credential.

Wrong Region (EU / Australia)

The CLI defaults to the US API. If the workspace is hosted in EU or Australia:

export INTERCOM_API_BASE_URL="https://api.eu.intercom.io"      # EU
export INTERCOM_API_BASE_URL="https://api.au.intercom.io"      # Australia

Set this in the shell rc file alongside the token.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.