AI Guide

agents
Marketing Agent Workflows on MCP: The Governance Layer Your Approval Log Is Missing AI Agent Development Services Powered by TeamAI How to Build an AI Agent Library: A Powerful Google Agentspace Alternative
AI Automation
Claude vs ChatGPT vs Gemini: 2026 Head-to-Head Comparison Understanding Gemini Models: A Plain-English Guide to Google's AI Family (2026) How to Automate Your Team's Workflows with AI: A Step-by-Step Guide AI Workspace for Teams: Benefits, Features & How to Choose One (2026) Best AI Models for Coding and Agentic Workflows in 2026 Best AI Models for Writing, Business Tasks and General Intelligence (2026) Who's Winning the AI Race in 2026? Claude vs ChatGPT vs Gemini in 2026: Giants, Challengers, and the AI model Showdown AI Model Benchmarks and Provider Comparison for 2026 22 AI Frontier Models Compared for 2026 How to Set Up AI Automated Workflows
AI Collaboration
The AI-Ready Team: How to Drive Adoption Without the Resistance How to Measure the ROI of AI Across Your Team AI Workspace for Teams: Benefits, Features & How to Choose One (2026) Best AI Models for Writing, Business Tasks and General Intelligence (2026) Who's Winning the AI Race in 2026? Claude vs ChatGPT vs Gemini in 2026: Giants, Challengers, and the AI model Showdown AI Model Benchmarks and Provider Comparison for 2026 22 AI Frontier Models Compared for 2026 How to Get My Team to Collaborate with ChatGPT
AI for Sales
Generating Sales Role-Play Scenarios with ChatGPT
AI Integration
Who's Winning the AI Race in 2026? Claude vs ChatGPT vs Gemini in 2026: Giants, Challengers, and the AI model Showdown AI Model Benchmarks and Provider Comparison for 2026 22 AI Frontier Models Compared for 2026 Integrating Generative AI Tools, like ChatGPT, into Your Team's Operations
AI Processes and Strategy
AI Overviews and AI Mode Cite Different Sources Most of the Time. Can a Prompt Library Track Both? AI Brief for AI Max: what each guideline type steers, and who should approve it How to Automate Your Team's Workflows with AI: A Step-by-Step Guide AI Workspace for Teams: Benefits, Features & How to Choose One (2026) Best AI Models for Writing, Business Tasks and General Intelligence (2026) How to Safeguard My Business Against Bad AI Use by Employees Providing Quality Assurance and Oversight of AI Like ChatGPT How to Choose the Right LLM for Your Business in 2026 How to Use ChatGPT & Generative AI to Scale a Team's Impact
Build an AI Agent
Creating a Custom AI Agent for Businesses Creating a Custom AI Marketing Agent Create an AI Agent for Sales Teams
Generative AI and Business
What Is the Cost of GEO in 2026? The 10 Top GEO Agencies for AI Visibility in 2026 Best AI Models for Writing, Business Tasks and General Intelligence (2026) The Benefits of AI for Small Businesses: Leveling the Playing Field Building a Data-Driven Culture With AI: A Practical Guide for Teams AI Terms Everyone Should Know (2026 Edition) Top 13 Alternatives to ChatGPT Teams Top 7 LLMs for Business in 2026: Ranked and Compared Will ChatGPT and LLMs Take My Job? Understanding the Value of ChatGPT and LLMs for Teams and Businesses Why Use ChatGPT & Generative AI for My Business
Large Language Models (LLMs)
Claude vs ChatGPT vs Gemini: 2026 Head-to-Head Comparison Understanding Gemini Models: A Plain-English Guide to Google's AI Family (2026) How to Automate Your Team's Workflows with AI: A Step-by-Step Guide AI Workspace for Teams: Benefits, Features & How to Choose One (2026) AI Model Economics: Choosing by Budget and Scale (2026) Best AI Models for Complex Reasoning Compared: Opus, GPT-5.4, Gemini, and Which to Use in 2026 Best AI Models for Coding and Agentic Workflows in 2026 Best AI Models for Writing, Business Tasks and General Intelligence (2026) Who's Winning the AI Race in 2026? Claude vs ChatGPT vs Gemini in 2026: Giants, Challengers, and the AI model Showdown AI Model Benchmarks and Provider Comparison for 2026 22 AI Frontier Models Compared for 2026 Every Gemini Model, Compared: Pricing, Context Windows & Which to Use DeepSeek R1, V4 Pro, and V4 Flash Compared: Pricing, Use Cases, and the Full 2026 Model Guide Every Claude Model, Compared: Versions, Pricing & Which to Use Every ChatGPT Model, Compared: Versions, Pricing & Which to Use Meet the Riskiest AI Models Ranked by Researchers Why You Should Use Multiple Large Language Models Overview of Large Language Models (LLMs)
LLM Pricing
How to Measure the ROI of AI Across Your Team AI Model Economics: Choosing by Budget and Scale (2026)
Prompt Libraries
How to Measure the ROI of AI Across Your Team How to Automate Your Team's Workflows with AI: A Step-by-Step Guide AI Prompt Templates for HR and Recruiting AI Prompt Templates for Marketers 8-Step Guide to Creating a Prompt for AI  What businesses need to know about prompt engineering How to Build and Refine a Prompt Library

Marketing Agent Workflows on MCP: The Governance Layer Your Approval Log Is Missing

TikTok launched Agentic Hub on June 30, 2026, a marketplace where advertisers browse ready-to-use first-party and third-party AI Skills covering campaign creation, creative generation, performance analysis, audience insight, and catalog management. TikTok’s launch announcement names publishers including HubSpot, Wix, Constant Contact, and Mobvista. The Hub is built on the TikTok for Business Model Context Protocol (MCP) server.

The connector and the marketplace are different layers, and you do not govern them the same way. TikTok now documents its MCP server as a bridge that packages approximately 400 TikTok API for Business endpoints as tools for campaign management, reporting, catalogs, creative, audiences, and Business Center administration.

TikTok says advertisers can use its MCP connection without handling API credentials, coding, or complex setup. TeamAI provides a user interface for registering an external MCP server as a workspace tool. Authentication requirements still depend on the server being connected, so some teams may need technical help with credentials and permission design. The harder question is what your record of agent actions needs to contain before any of those tools can touch live budget.

Four layers sit between an agent and your ad budget

Most write-ups treat MCP as a single thing. It is four, and knowing which layer you are looking at determines where your log has to sit.

Click a layer to see what it means for governance
Agentic Hub
The marketplace where Skills are discovered and adopted
Adoption is a permission decision, not an installation detail.
TikTok Ads Skills
Ready-to-use jobs such as campaign creation, performance analysis, and creative diagnostics
A broad Skill request may produce more than one account change, so do not treat one request as one action.
TikTok for Business MCP server
Packages approximately 400 TikTok API for Business endpoints as tools an MCP client can call
This is the layer where concrete account actions happen, so log the resulting actions.
TikTok Marketing API
The underlying account, campaign, creative, audience, and reporting surface
Defines the outer limit of what any agent can ever do.
Ordered top to bottom: discovery layer down to the foundation layer.

The third row is the one that catches teams out. If you log only the Skill request, one entry can hide several distinct changes to campaigns, ad groups, ads, or budgets. Log each resulting account action and reconcile those actions back to the request.

TeamAI sits above all four as the orchestration layer. Your agent, its instructions, its grounded knowledge, and its registered tools live in the workspace, and you can register an external MCP server as a workspace tool from Tools, then Create Tool, then the MCP Server type. TeamAI is not TikTok Ads Manager and does not replace it. TikTok Ads Manager and TikTok API for Business stay the systems of record, and TikTok’s own permission model still decides what any call is allowed to do. What the workspace gives you is one governed place where the agent, the connection, and the human review step sit together instead of scattered across personal accounts.

The read and write taxonomy that should drive your scope

Nearly every credible piece on this topic lands on a read versus write distinction. That distinction is correct and too coarse to scope a permission against. Four classes are more useful.

Click a class to see its example and default posture
Lower riskHigher risk
Example
Pulling a performance report, summarizing which campaigns underperformed
Default posture
Allow early, allow broadly

The class teams misjudge is the third one. Reversibility is not a property the platform grants you. It is a property you create by capturing the prior value before the call executes. If nobody recorded what the budget was at 9:14 that morning, the change was not reversible, whatever the interface suggests. Write that down as a rule before you write it down as a field.

Google’s official Analytics MCP server is a clean example of the first class and a useful place to start a read-only connection. Its credentials require the analytics.readonly scope, so the connection does not receive Analytics write permission. Google labels the server Experimental, and the repository describes it as a local server, so setup requires local tooling and Google credentials rather than a hosted no-code connection.

A public tool catalog does not replace outcome verification

TikTok now publishes an extensive MCP tool catalog and a changelog. That closes the documentation gap. It does not remove the need to verify what happened inside the ad account.

An accepted tool call proves that the platform processed a request. It does not prove that the resulting object, value, or delivery state matches what the approver intended. Keep two fields in the record.

First, record intended state and observed state as separate fields, permanently, not just during a pilot. Intended state is what your workflow asked for. Observed state is what you confirmed in TikTok Ads Manager afterwards. When those two disagree, you want the disagreement visible in the record rather than reconstructed from memory.

Second, hand-verify the first several writes of every new Skill in TikTok Ads Manager before you rely on it for recurring actions. Repeat that verification after a relevant tool or platform update. This is outcome validation, not a substitute for documentation.

Least-privilege scope with the controls that actually exist

Scoping advice usually stops at “use minimum permissions.” The useful version names the specific controls and, more importantly, what each one does not cover.

Click a control to see what it constrains, and what it does not
Registering the MCP server as a workspace tool, with the Authorization header set to Hidden
Constrains
Where the connection lives and whether the token is stored securely rather than pasted in plain text.
Does not constrain
What the token itself is permitted to do on the far side.
Tool and agent visibility, set to private, workspace-wide, or organization-wide
Constrains
Who inside your organization can see and run the connection.
Does not constrain
Whether a permitted runner should be making live budget changes.
Roles and Permissions under Workspace Settings
Constrains
Who can create and edit agents and tools. TeamAI states that owners can currently restrict this, and that additional permission controls are rolling out over time.
Does not constrain
Per-action approval on an individual write. That discipline has to live in your workflow design and in the token you issue.
Organization API keys on Enterprise, scoped to All Workspaces or to Specific Workspaces
Constrains
Which workspaces a key can reach. Choose Specific Workspaces.
Does not constrain
What the connected platform permits within those workspaces.
Plan gates: Workspace MCP integration on Professional and Enterprise, plus legacy Starter and Best of Breed plans
Constrains
Who can stand a connection up at all. Organization integration requires Enterprise, and an Organization Admin or Owner role to create keys.
Does not constrain
How carefully it gets reviewed afterwards.

Read the right-hand column twice. It is the honest shape of this problem: the workspace can govern who connects what and where the credential lives, and it cannot govern whether a change was a good idea. That judgment stays with a person, and your ledger is how you prove a person made it.

One more honest note from TeamAI’s own documentation: remote MCP connections are still evolving technology. The setup guidance recommends restarting the client or temporarily disabling and re-enabling the server if a connection fails. Build your first connection when you have time to test it, not an hour before a campaign launch.

What a per-action ledger should actually contain

“Keep an approval log” is advice nobody can execute. Here are the fields.

Filter the ledger by field group
action_id
One row per resulting account action, not per Skill request
Why it matters: One request may affect several account objects or values
connection_id
Which registered MCP tool made the call
Why it matters: Lets you retire one connection without auditing all of them
token_scope
Specific Workspaces or All Workspaces
Why it matters: The most common over-permission, visible at a glance
surface
TikTok Ads via a named Skill, GA4 read-only, internal knowledge
Why it matters: Separates spend-touching calls from analysis
action_class
Read, draft-only, reversible live, irreversible live
Why it matters: Drives which review and cadence rules apply
intended_state
Active, paused, or draft, as the workflow asked for it
Why it matters: Records what the approver expected the platform to produce
observed_state
What you confirmed in the platform interface
Why it matters: The other half. Blank means unverified, not fine
prior_value
The value before the change, or an explicit \u201cnot captured\u201d
Why it matters: Without this, reversible is a claim you cannot support
approval_status
Approved before execution, notified after execution, or unreviewed
Why it matters: Three states, never two
approver
A named person
Why it matters: \u201cThe team\u201d is not an approver
approved_at
Timestamp of the human decision
Why it matters: Deliberately separate from the next field
executed_at
Timestamp of the agent action
Why it matters: If this precedes approved_at, the workflow acts first and tells you later
reversal_path
The exact steps to undo, or \u201cnone\u201d
Why it matters: Written before the action, not improvised after it
recheck_due
Next scheduled review of this connection
Why it matters: Turns a launch decision into a maintained one

Two of these get dropped almost every time, and they are the two that carry the audit.

prior_value is the first. It is the difference between a change you can undo and a change you can only describe.

The approved_at and executed_at split is the second. Keeping one timestamp lets “a human reviewed this” and “a human was told afterwards” collapse into the same row, and once they collapse you can no longer tell which kind of workflow you built.

Approved before is not the same status as notified after

Three approval statuses, not two.

Approved before execution means a person released the action. Notified after execution means the action happened and a person can now react. Unreviewed means nobody looked. These carry genuinely different risk, and a log that records the second as “reviewed” is worse than no log, because it produces false confidence in an audit.

If the honest answer for a given workflow is notified after execution, record that. Then decide whether the action class justifies it. Reads and draft-only writes usually do. Live budget changes usually do not.

TeamAI’s Automated Workflows are where the before-approval pattern gets built: chain the analysis steps, route the output for a human decision, and place the write call behind that step rather than in front of it. The draft-then-review shape is the same idea at the content layer, where the model’s output stays inside the workspace until a person releases it. For API-type connections and the plugin builder specifically, Custom Plugins covers the setup path this page does not repeat. The workspace can route your approval, but the connected platform’s permission system remains the final limit on what the credential can do.

Multi-model routing helps on the analysis half of these workflows, where the TeamAI FAQ lists access to more than twenty leading models across providers, so the reasoning step and the drafting step do not have to run on the same model. Routing has no bearing on write safety, and it is worth keeping those two decisions separate in your own head.

A recheck cadence tied to action class, not the calendar

A quarterly review of everything is how drift survives. Tie the cadence to what the connection can do.

Reviews per year, by action class
Hover a bar for what you verify at that cadence

Two triggers override this schedule: adding a new Skill, agent, or token to an existing workflow, and any relevant platform or MCP changelog update. Both call for an immediate re-review, not a wait for the next cycle.

Two triggers override the schedule. Add a new Skill, agent, or token to an existing workflow and re-review immediately rather than waiting for the cycle. Re-verify after any relevant platform or MCP changelog update, because tool behavior and supported actions can change.

The failure pattern this prevents is boring, which is why it works. A team wires a write-enabled agent, scopes it carefully, reviews the first few actions closely, and then the review slips as other priorities take over. The agent keeps acting inside its permitted scope, so nothing looks broken in any single moment. Three months later nobody can say what changed across that stretch. The scope was never the problem. The maintained record was.

Where this stops being a workflow problem

If you are governing one channel, build this yourself. The ledger above is a spreadsheet, the read-only and draft-then-review tiers are worth wiring regardless of what you decide about live writes, and TeamAI gives you one workspace where the agent, the connection, the credential, and the approval step sit together instead of in four places. Plan gates for MCP connections are on pricing. TeamAI’s enterprise page says WebFX uses the workspace across its organization of more than 500 employees and includes a CEO statement that hundreds of employees share prompts and models there.

If you are governing several channels, the harder question is a different one. The cited Google, Meta, and TikTok product documentation describes each platform’s own capabilities. It does not establish one shared authority that adjudicates budget decisions across all three. That is a coordination problem rather than a permission problem, and it is worth reading who is coordinating agent decisions across your ad platforms before you scale a workflow past one account.

One fork worth naming plainly. If what you actually want is somebody else accountable for paid media outcomes rather than a workspace your own team governs, that is Growth Loops, the managed paid marketing program on teamai.com. Different product, different buyer job, and not what this page is describing.

Ready to build the governed version? Start in TeamAI, register your first read-only MCP connection, and put the ledger in place before you ever grant a write.

Start in TeamAI