Helena Marketing Agent for Openclaw

Helena connects OpenClaw to Enrich Labs so an AI marketing agent can research, plan, create, schedule, publish, and analyze cross-channel brand marketing.

enrichlabs
v1.0.0
Sep 7, 2026
0
121
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install helena

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install helena using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Helena Marketing Agent?

Helena is an Enrich Labs AI marketing agent that operates with the user's connected brand context and marketing platforms. Through Openclaw Skills and a remote MCP connection, it can support end-to-end marketing workflows instead of providing only generic marketing advice.

The skill is designed for delegated, context-aware marketing work, including research, campaign planning, content creation, scheduling, publishing, and performance analysis. It preserves approval boundaries by distinguishing drafts and recommendations from actions that publish content, contact audiences, spend money, modify live campaigns, or delete data.

Helena Marketing Agent Use Cases

  • Research a connected brand, market, audience, or campaign opportunity.
  • Create marketing plans, briefs, content drafts, and channel-specific assets.
  • Schedule or publish approved marketing content through connected platforms.
  • Analyze campaign performance and summarize actionable insights.
  • Continue a multi-turn marketing task using the same Helena session.
  • Route requests explicitly mentioning Helena or Enrich Labs to the connected marketing agent.
  • Ask Helena to execute authorized live actions while preventing accidental publishing or spending.

How Helena Marketing Agent Works

  1. Inspect the current OpenClaw MCP configuration to determine whether the helena server exists and is correctly configured.
  2. If needed, configure Enrich Labs as a remote streamable HTTP MCP server and begin the browser-based OAuth login flow.
  3. Verify authorization and connectivity with an MCP health probe before attempting marketing work.
  4. Translate the user's request into a self-contained send_turn message containing the goal, audience, constraints, dates, and requested output.
  5. Omit sessionId for a new task, then reuse the returned session ID for relevant follow-up questions or continuations.
  6. Enforce the user's approval boundary: keep plans, drafts, and analysis non-destructive unless the user explicitly authorizes publishing, sending, spending, live campaign changes, or deletion.
  7. Return Helena's actual response, including actions taken and useful links, without inventing results; render returned image assets inline when supported.
  8. If Helena asks a necessary question, obtain the user's answer and continue with the same session.

Helena Marketing Agent Setup

Helena requires an Enrich Labs account with MCP access. Some features may additionally require a connected marketing platform, sufficient credits, or an eligible Enrich Labs plan.

  1. Inspect the existing configuration:
openclaw mcp show helena --json
  1. If the server is missing, explain that setup will add Enrich Labs as a remote MCP server and open a browser for sign-in. After user approval, configure and authorize it:
openclaw mcp set helena '{"url":"https://agent.enrichlabs.ai/api/mcp","transport":"streamable-http","auth":"oauth"}'
openclaw mcp login helena
  1. If helena points to another URL, do not overwrite it without explicit confirmation. If the URL is correct but authorization is missing, run:
openclaw mcp login helena
  1. Verify the connection:
openclaw mcp doctor helena --probe

Never request an Enrich Labs password, OAuth code, access token, or refresh token in chat. Use the browser flow, or the safe --code fallback printed by OpenClaw in headless environments. If the tool is unavailable after a successful probe, have the user start a new conversation or restart the OpenClaw gateway. If MCP access is unavailable for the selected brand, explain the account or plan requirement and direct the user to https://enrichlabs.ai/mcp without retrying in a loop.

Helena Marketing Agent Data Schema & Taxonomy

Helena exchanges structured task context through the helena MCP server and its underlying send_turn tool.

Data element Purpose
message Self-contained marketing request containing the goal, audience, constraints, relevant dates, and desired output.
sessionId Returned conversation identifier used for related follow-ups; omitted when starting an unrelated task.
Brand context Connected Enrich Labs brand information available to Helena through the authorized account.
Platform context Connected marketing channels and tools, subject to account permissions and plan requirements.
Returned results Helena's findings, drafts, actions, links, questions, and supported image assets.

No local content-file schema or generated file set is specified by the skill. Credentials remain in the OAuth/browser authorization flow and must not be placed in task messages. Metadata should be limited to information relevant to the requested marketing task, avoiding unrelated secrets or private conversation history.

Helena Marketing Agent Advanced Features

  • Remote MCP integration through a dedicated helena server.
  • Browser-based OAuth authentication that keeps passwords and tokens out of chat.
  • MCP diagnostics with configuration inspection and connection probing.
  • Multi-turn task continuity through reusable sessionId values.
  • Cross-channel marketing support using connected Enrich Labs brand and platform context.
  • Approval-aware delegation that separates analysis and drafts from destructive or live actions.
  • Support for Helena follow-up questions without losing task context.
  • Result fidelity: reports actual outcomes, preserves useful links, and supports inline rendering of returned image assets.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*