← Back to Skills Marketplace
rmorse

Help Scout

by Ross Morsali · GitHub ↗ · v0.0.2
cross-platform ✓ Security Clean
306
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install helpscout-cli
Description
CLI for the HelpScout API. Manage conversations, customers, mailboxes, knowledge base articles, and more from the terminal. Covers both Inbox and Docs APIs w...
README (SKILL.md)

hs

Use hs to interact with HelpScout from the terminal. Two API namespaces: hs inbox (Mailbox API — conversations, customers, users, teams, etc.) and hs docs (Docs API — sites, collections, categories, articles).

Auth

Inbox uses OAuth2 client credentials (App ID + App Secret). Docs uses an API key.

  • hs inbox auth login — interactive setup, validates against the API
  • hs docs auth login — prompt for Docs API key, validates
  • hs inbox auth status / hs docs auth status — check stored credentials
  • hs inbox auth logout / hs docs auth logout — remove credentials

Credential resolution order: OS keyring → config file. For non-interactive auth, use hs inbox config set --inbox-app-id \x3Cid> --inbox-app-secret \x3Csecret>.

Inbox commands

Conversations (conv)

  • hs inbox conv list --status active --mailbox \x3Cid> --tag \x3Cname> --assigned-to \x3Cuid> --embed threads
  • hs inbox conv get \x3Cid> --embed threads
  • hs inbox conv create --mailbox \x3Cid> --subject "..." --customer \x3Cemail> --body "..."
  • hs inbox conv update \x3Cid> --status closed --subject "..."
  • hs inbox conv delete \x3Cid>
  • hs inbox conv threads list \x3Cconv-id>
  • hs inbox conv threads reply \x3Cconv-id> --customer \x3Cemail> --body "..." --status closed
  • hs inbox conv threads note \x3Cconv-id> --body "..."
  • hs inbox conv tags set \x3Cconv-id> --tag billing --tag urgent
  • hs inbox conv fields set \x3Cconv-id> --field \x3Cid>=\x3Cvalue>
  • hs inbox conv attachments upload \x3Cconv-id> --thread-id \x3Cid> --file ./path

Customers (cust)

  • hs inbox cust list --query "email:[email protected]"
  • hs inbox cust get \x3Cid>
  • hs inbox cust create --first-name Jane --last-name Doe --email [email protected]
  • hs inbox cust update \x3Cid> --last-name Smith
  • hs inbox cust delete \x3Cid>

Mailboxes (mb)

  • hs inbox mb list / hs inbox mb get \x3Cid>
  • hs inbox mb folders list \x3Cmb-id> / hs inbox mb custom-fields list \x3Cmb-id>

Users

  • hs inbox users list --email [email protected] / hs inbox users get \x3Cid> / hs inbox users me
  • hs inbox users status get \x3Cuid> / hs inbox users status set \x3Cuid> --status away

Teams

  • hs inbox teams list / hs inbox teams members \x3Cteam-id>

Organizations

  • hs inbox organizations list --query "acme" / hs inbox organizations get \x3Cid>
  • hs inbox organizations create --name "Acme Corp" / hs inbox organizations delete \x3Cid>

Tags / Ratings

  • hs inbox tags list / hs inbox ratings list

Workflows (wf)

  • hs inbox wf list / hs inbox wf run \x3Cid> --conversation-ids id1,id2
  • hs inbox wf update-status \x3Cid> --status active

Webhooks (wh)

  • hs inbox wh list / hs inbox wh get \x3Cid>
  • hs inbox wh create --url https://... --events convo.created --secret s3cret
  • hs inbox wh delete \x3Cid>

Saved replies

  • hs inbox saved-replies list --mailbox-id \x3Cid>
  • hs inbox saved-replies create --mailbox-id \x3Cid> --name "Greeting" --body "Hello..."

Reports

  • hs inbox reports conversations --start 2025-01-01 --end 2025-01-31 --mailbox \x3Cid>
  • Subcommands: chats, company, conversations, customers, docs, email, productivity, ratings, users

Tools

  • hs inbox tools briefing — daily briefing summary
  • hs inbox tools briefing --assigned-to \x3Cuid> --embed threads — agent-specific briefing with thread content

Docs commands

Sites

  • hs docs sites list / hs docs sites get \x3Cid>
  • hs docs sites create --subdomain help --title "Help Center"

Collections

  • hs docs collections list --site \x3Cid> / hs docs collections get \x3Cid>
  • hs docs collections create --site \x3Cid> --name "Getting Started"

Categories

  • hs docs categories list \x3Ccollection-id> / hs docs categories get \x3Cid>
  • hs docs categories create --collection \x3Cid> --name "FAQ"
  • hs docs categories reorder \x3Ccollection-id> --categories id1,id2,id3

Articles

  • hs docs articles list --collection \x3Cid> or --category \x3Cid>
  • hs docs articles search --query "password reset" --site \x3Cid>
  • hs docs articles get \x3Cid> / hs docs articles get \x3Cid> --draft
  • hs docs articles create --collection \x3Cid> --name "How to reset" --text "Step 1..."
  • hs docs articles update \x3Cid> --text "Updated..." --status published
  • hs docs articles delete \x3Cid>
  • hs docs articles draft save \x3Cid> --text "..." / hs docs articles draft delete \x3Cid>
  • hs docs articles revisions list \x3Cid> / hs docs articles revisions get \x3Cid> \x3Crev-id>
  • hs docs articles upload \x3Cid> --file ./image.png

Redirects

  • hs docs redirects list \x3Csite-id> / hs docs redirects find --site \x3Cid> --url /old-path
  • hs docs redirects create --site \x3Cid> --url-mapping /old --redirect /new

Assets

  • hs docs assets article upload --file ./img.png
  • hs docs assets settings upload --file ./logo.png

Output & global flags

  • --format table|json|json-full|csv — output format (default: table)
  • --no-paginate — fetch all pages automatically
  • --page \x3Cn> / --per-page \x3Cn> — pagination (defaults: 1, 25)
  • --debug — show HTTP request/response details

Config

  • hs config set --format json --inbox-default-mailbox 12345
  • hs config get [key] — print one or all config values
  • hs config path — print config file location
  • Default location: ~/.config/hs/config.yaml (Linux/macOS), %AppData%\hs\config.yaml (Windows)

PII redaction

Inbox commands redact customer/user PII with deterministic fake identities. Controlled via config key inbox_pii_mode.

Mode Effect
off No redaction (default)
customers Redact customer names, emails, phones
all Redact both customers and users
  • --unredacted flag on hs inbox disables redaction for one call (requires inbox_pii_allow_unredacted: true in config)

Permissions

Allowlist model — restrict which commands an agent can run. Empty policy = unrestricted.

  • Set via config key inbox_permissions / docs_permissions — comma-separated resource:operation pairs
  • Wildcards: *:read, conversations:*
  • Operations: read, write, delete, *
  • hs inbox permissions — inspect current policy, shows ALLOW/DENY per command

Example: hs inbox config set --inbox-permissions "conversations:read,customers:read,mailboxes:read"

Notes

  • Self-update: hs update
  • Shell completions: hs completion bash|zsh|fish|powershell
Usage Guidance
Before installing: 1) Verify the Homebrew tap operator-kit/tap and inspect its formula/source on GitHub to ensure you trust the provider. 2) Understand that using the skill means the hs CLI will store and read Help Scout credentials from your OS keyring and the config file (~/.config/hs/config.yaml); treat those as sensitive. 3) Be cautious when running commands that upload local files or produce briefings containing conversation text (may include PII) — avoid pointing the CLI at sensitive files unless necessary. 4) If you prefer more control, install the CLI manually, review its binary/source, and perform interactive auth rather than providing credentials via automation. 5) If you have low tolerance for third-party packages, consider obtaining the client from an official Help Scout source or a vetted package repository.
Capability Analysis
Type: OpenClaw Skill Name: help-scout Version: 0.0.2 The skill bundle describes a legitimate CLI tool for interacting with the HelpScout API. The `SKILL.md` file provides clear, descriptive instructions for using the `hs` command-line interface, including authentication, various CRUD operations for HelpScout resources, and configuration options. There are no indications of prompt injection attempts against the AI agent, no instructions for unauthorized data exfiltration, persistence, or malicious execution. The installation method via `brew` is standard. While the underlying `hs` tool has capabilities like file uploads and webhook creation that *could* be misused by a compromised agent, the skill bundle itself does not instruct or encourage such misuse, nor does it contain any hidden malicious logic.
Capability Assessment
✓ Purpose & Capability
Name/description match the declared requirement for the 'hs' CLI binary and the SKILL.md documents CLI commands for the Inbox and Docs APIs. The required binary (hs) and the brew install formula align with the stated purpose.
ℹ Instruction Scope
Instructions are focused on using the hs CLI and describe auth flows, config location (~/.config/hs/config.yaml), and operations (conversations, docs, uploads, webhooks, etc.). They do direct the agent to operate on local files when uploading attachments/assets and to read/store credentials in the OS keyring and config file. That is expected for a CLI but means the agent may access local files you point it at (attachments) and local secrets in the keyring/config.
ℹ Install Mechanism
Install uses a Homebrew formula (operator-kit/tap/hs) which is a package-managed install (lower risk than arbitrary downloads). However, the formula is from a third-party tap (operator-kit/tap) rather than the main homebrew/core; you should verify the tap/source before installing.
ℹ Credentials
The skill declares no required environment variables, which is consistent. The SKILL.md does rely on storing and using Help Scout credentials (OAuth client id/secret and Docs API key) in the OS keyring or config file; those are proportionate to a Help Scout CLI but are sensitive and will be read/written by the CLI as part of normal operation.
✓ Persistence & Privilege
The skill does not request always:true or elevated platform privileges. It will install a binary and create/modify its own config and keyring entries (expected behavior for a CLI). The agent's ability to invoke the skill autonomously is the platform default and not by itself a red flag.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install helpscout-cli
  3. After installation, invoke the skill by name or use /helpscout-cli
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.0.2
- Updated authentication flow: environment variables are no longer used or documented for auth; now credentials resolve via OS keyring or config file. - Changed guidance for non-interactive authentication to use `hs inbox config set` command. - Removed documentation for configuration and control via environment variables. - Permission and redaction settings are now managed via config keys (`inbox_permissions`, `docs_permissions`, `inbox_pii_mode`) instead of environment variables. - Updated instructions and examples throughout SKILL.md for clarity and consistency with the new config-driven approach.
v0.0.1
**Initial version uses HS CLI** - Wrapper for the open-source `hs` CLI. - Manages all HelpScout Inbox and Docs resources via CLI: conversations, customers, mailboxes, users, knowledge base, etc. - Adds new features such as PII redaction, granular permissions, and multiple output formats. - Updates metadata and setup to require the `hs` binary and provides install instructions.
Metadata
Slug helpscout-cli
Version 0.0.2
License —
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is Help Scout?

CLI for the HelpScout API. Manage conversations, customers, mailboxes, knowledge base articles, and more from the terminal. Covers both Inbox and Docs APIs w... It is an AI Agent Skill for Claude Code / OpenClaw, with 306 downloads so far.

How do I install Help Scout?

Run "/install helpscout-cli" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Help Scout free?

Yes, Help Scout is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Help Scout support?

Help Scout is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Help Scout?

It is built and maintained by Ross Morsali (@rmorse); the current version is v0.0.2.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments