Marketing Engineer turns marketing operations into an evidence-based, auditable workflow with structured workspaces, content tickets, GEO/SEO playbooks, and a live dashboard.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install marketing-engineer
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install marketing-engineer 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 Marketing Engineer?
Marketing Engineer is one of the most comprehensive Openclaw Skills for organizing product marketing as an engineering system: knowledge becomes vertical strategy, strategy becomes work tickets, content moves through review gates, and performance data feeds back into the workspace. It helps teams produce consistent, source-backed marketing assets instead of relying on disconnected ideas or undocumented claims.
The skill creates a single-file marketing dashboard that provides a workspace overview, directory tree, file statuses, calendars, countdowns, ticket views, channel records, and decision areas. It also supports content planning, keyword research, GEO and SEO optimization, AI search visibility diagnosis, multi-channel distribution, competitor intelligence, and campaign operations while keeping human approval in the publishing loop.
Marketing Engineer Use Cases
- Create a repeatable marketing or growth workspace for a product, startup, or campaign.
- Review the complete state of a workspace, including directories, document status, deadlines, and upcoming publishing activity.
- Plan and produce content for channels such as blogs, newsletters, Zhihu, Xiaohongshu, X, Medium, and other owned or social platforms.
- Conduct keyword research, SEO planning, GEO optimization, and AI search visibility diagnostics.
- Research competitors, market trends, audience discussions, and cross-platform signals.
- Convert marketing briefs into structured tickets with blockers, review steps, publication details, and measurable outcomes.
- Maintain consistent brand terminology, domains, pricing language, invitation codes, and product positioning across channels.
- Manage launch campaigns, channel matrices, distribution plans, publishing records, and performance feedback.
- Use optional Openclaw Skills for research, challenge-based planning, specification generation, and ticket decomposition without changing the approval gates.
How Marketing Engineer Works
- Initialize the workspace: Create a new marketing workspace or connect the dashboard to an existing directory.
- Populate shared knowledge: Record verified product facts, positioning, approved terminology, channel information, sources, dates, and toolchain metadata before producing content.
- Research before writing: Gather missing evidence through an available research skill or WebSearch/WebFetch, and label unsupported claims as
[Unverified]until validated. - Create a brief and ticket: Define the audience, objective, evidence, risks, deliverable, and blockers, then move the work through the structured content-production pipeline.
- Draft and review: Produce the draft using the shared knowledge base, apply the relevant channel and GEO/SEO rules, and send it for human review.
- Finalize and publish: Apply layout or creative assets only after approval, publish through the selected channel, and record the actual live version in the publishing ledger.
- Audit consistency: Scan terminology, product claims, domains, pricing language, and internal views whenever positioning changes.
- Refresh the dashboard: Scan the workspace and rebuild or serve the dashboard so its tree, file states, calendar, tickets, and reporting views reflect the current source files.
- Close the loop: Add publication URLs, outcomes, and performance data to the appropriate records so future planning benefits from documented results.
Marketing Engineer Setup
- Ensure Python 3 is available and place the skill in the agent's installed skills directory.
- Read the primary operating guide before using the skill:
cat references/usage-manual.md
- Create a new workspace with the optional dashboard:
python3 scripts/init_workspace.py <workspace-directory> --product "<product-name>" --with-dashboard
- For an existing workspace, generate or refresh the dashboard:
python3 scripts/build_dashboard.py <workspace-directory>
python3 scripts/build_dashboard.py <workspace-directory> --force
- Start the local dashboard server when real-time scanning is needed:
python3 scripts/serve_dashboard.py <workspace-directory> 8799
Open http://127.0.0.1:8799/dashboard.html in a browser. The server exposes workspace tree and file-reading endpoints for live views.
- Complete the shared knowledge files according to
references/workspace-structure.md, then followreferences/workflow-content-production.md,references/geo-playbook.md, andreferences/brand-consistency-audit.mdas needed. - Store API keys only in
.env; never place secrets in content, reports, dashboard files, or source documents. Optional browser-cookie access must remain disabled unless the user explicitly authorizes it.
Marketing Engineer Data Schema & Taxonomy
The workspace is organized as a source-controlled marketing knowledge and production system:
| Area | Purpose and expected data |
|---|---|
shared-knowledge/ |
Verified product facts, approved positioning, terminology, channel matrix, evidence sources, dates, and toolchain.md metadata. This is the primary source for content generation. |
| Briefs and tickets | Planning records that define objectives, audience, evidence, risks, blockers, deliverables, owners, and review status. |
| Drafts and published assets | Content organized according to the workspace directory conventions, with separate draft, approved, finalized, and published states. |
results/channel-posts.md |
Actual publication records, including channel, live URL, publication details, and the version that was really posted. |
| Research and audits | Markdown records containing source URLs, dates, findings, verification labels, GEO/SEO observations, and brand-consistency results. |
.cache/workspace_map.json |
Generated workspace inventory containing directory meanings, file dates, statuses, and descriptions. |
dashboard.html |
Generated single-file visual view derived from workspace sources; it is a presentation layer, not the authoritative data source. |
| Calendar and campaign records | Deadlines, upcoming activities, countdowns, launch milestones, and optional calendar synchronization artifacts such as .ics. |
| FAQ and structured data | Markdown FAQ drafts converted by scripts/md-faq-to-schema.py into schema.org FAQPage JSON-LD, with visible text kept consistent with the generated markup. |
The taxonomy emphasizes evidence provenance, verification state, content lifecycle, channel, publication status, deadline, and feedback loop. Facts without a source and date are marked [Unverified] and must not enter finished content. Dashboard views are regenerated from tickets, ledgers, channel tables, and reports rather than treated as the source of truth.
Marketing Engineer Advanced Features
- Single-file marketing dashboard with workspace-wide overview, two-column directory and file views, four-level status evaluation, live scanning, Markdown rendering, image previews, code copying, and light/dark themes.
- Local dashboard server with
/api/treereal-time workspace scanning and/api/filefull-file retrieval. - Automatic dashboard backups and offline text snapshots during generation.
- Calendar, deadline, countdown, campaign, ticket, ledger, and decision-area views in one operational interface.
- Optional Apple Calendar synchronization with
.icsand open-file fallback when AppleScript permissions are blocked. - GEO playbook covering five-step keyword research, AI search source mapping, recurring probes, visibility baselines, and KPIs.
- Brand-consistency audits that scan approved terminology, domains, free-plan language, invitation codes, internal views, and archived findings.
- Optional integration with
last30days,grilling,to-spec,to-tickets, andresearchOpenclaw Skills, with documented fallback paths when those skills are unavailable. - Human-in-the-loop publishing gates: external skills can accelerate research or drafting, but they cannot bypass review or redefine the publishing ledger.
- Privacy-first handling of credentials, browser cookies, and local data, with explicit authorization required for sensitive access.
- FAQ Markdown to schema.org FAQPage JSON-LD conversion for search-friendly pages while preserving consistency between structured data and visible copy.
SKILL.md
Loading
Related Openclaw Skills
forktree
forktree helps coding agents and automation create, manage, and clean up Git worktrees with safety checks and predictable shell output.
Medical Device Registration Dossier Assistant
A regulatory writing assistant that organizes medical device evidence into market-specific, submission-ready registration dossiers.
Market & Customer Insight
A structured market and customer insight workflow that combines ST, APPEALS, and KANO analysis to identify high-value segments, unmet needs, and blue-ocean opportunities.
Xiaoyunque Short Drama Creator
An Openclaw Skills workflow for submitting, monitoring, and downloading AI-generated short drama projects through Pippit Tool CLI.
Learning5D Agent Safety Check and Enrolment
Learning5D helps agents assess safety risks, review skills before installation, and optionally study consent-first relational safety.
AUTO.Prompt
AUTO.Prompt guides an AI agent to plan before acting, challenge its plan once, execute the shortest suitable path, and save only reusable knowledge.
Featured*
