cmo-studio-integration

Narrates marketing agent actions into the mktg-studio dashboard via localhost HTTP API calls.

31|5|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/MoizIbnYousaf/marketing-cli --skill cmo-studio-integration-moizibnyousaf
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cmo-studio-integration
Source: https://github.com/MoizIbnYousaf/marketing-cli/tree/main/studio/skills/cmo-studio-integration
Command: npx skills add https://github.com/MoizIbnYousaf/marketing-cli --skill cmo-studio-integration-moizibnyousaf

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When the mktg-studio dashboard is running locally, the /cmo marketing agent works invisibly — the user cannot see which skills ran, which brand files changed, or what to do next. This Skill makes the agent proactively report every meaningful action to the dashboard's localhost HTTP API so the user watches work happen in real time. ## Core Features & Use Cases - Health Detection: Pings /api/health once per turn to detect whether the studio is running, and silently falls back to CLI-only behavior when it is down. - Five Reporting Verbs: Logs skill runs via /api/activity/log, notes brand-file edits via /api/brand/note, pushes recommended next actions via /api/opportunities/push, shows toasts via /api/toast, and switches dashboard tabs via /api/navigate. - Atomic Brand Writes: Uses /api/brand/read and /api/brand/write with expectedMtime optimistic locking so concurrent user edits return a 409 conflict instead of silent overwrites. - Use Case: After running a landscape-scan skill, the agent logs the run, notes the updated brand/landscape.md, pushes a follow-up opportunity, navigates the user to the Signals tab, and fires a success toast — all with eight curl calls. ## Quick Start Ask the agent to check whether mktg-studio is running and log its next marketing action to the dashboard activity feed.

Frequently Asked Questions about cmo-studio-integration

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make an AI agent report its actions to a local dashboard?▼

Detect the dashboard with a curl health check against /api/health, then POST each action to endpoints like /api/activity/log and /api/toast. Each call is fire-and-forget and the dashboard updates in real time via SSE.

How to safely write files through an HTTP API with optimistic locking?▼

Read the file via /api/brand/read to capture its mtime, then POST to /api/brand/write with expectedMtime. A concurrent edit returns 409 CONFLICT with a fix hint pointing back to re-read and merge.

What happens when the mktg-studio dashboard is not running?▼

The health check fails and the agent behaves exactly as in a CLI-only setup. It does not retry, surface errors, or mention the unreachable studio to the user.

Which mktg-studio API endpoints do not return the standard ok envelope?▼

Three GET endpoints return bare objects without the ok/data wrapper: /api/opportunities, /api/audience/profiles, and /api/briefs. Treat a 200 with the expected top-level keys as success.

When should I avoid sending navigate or toast events?▼

Send one navigate per logical destination change and one toast per user-facing milestone. Routine progress belongs in /api/activity/log; spamming navigate or toast between every step of a multi-skill playbook creates noise.