AdsAgent — Google Ads MCP

approval_plan_confirm

Destructive

After explicit user approval of every displayed item, execute the stored mixed advertising plan identified by approval_ref and expected_plan_digest. Supply the exact expected_changes returned by prepare so the host can inspect targets and effects; the server rejects any mismatch. May create, copy, edit, remove or change delivery, budgets and bids of advertising resources, affecting spend. Only registered operation-specific executors run, with their current permissions and validation. Execution is sequential, not atomic at the provider. Stop on failure or uncertainty; preserve per-item receipts. A repeat returns the existing outcome or resumes only proven-unsent quota-deferred items after retry_at within the original approval expiry. Present one approval for the complete requested change set within this platform. Show account and target names, operation, before/after values, warnings and expiry. Keep approval refs, draft IDs, digests, tokens and protocol JSON in tool traffic, not normal chat. Ask for one confirmation of the whole displayed plan. Report each completed, submitted, waiting, failed or uncertain item accurately. A submitted background task is not a verified provider change. Never replay an uncertain write or retry a host safety denial. A waiting_quota plan retains its confirmation only until its original expiry; resume the exact plan after retry_at without asking for another confirmation. Changed or expired plans require review.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`approval_ref`	Yes
`expected_changes`	Yes
`connection_context`	No
`expected_plan_digest`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 13, 2026
- addedInput schema / properties / connection_context
```
Added value: +{
+  "pattern": "^mctx_v1_[0-9a-f]{64}$",
+  "type": "string"
+}
```
AddedOct 12, 2026

TDQS

A4/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the destructiveHint/openWorldHint annotations, it discloses a lot: sequential non-atomic execution at the provider, stop-on-failure behavior, per-item receipts, idempotent repeat semantics with quota-deferred resume after retry_at bounded by approval expiry, and that only registered executors with current permissions run. This is unusually rich behavioral context for a mutation tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness2/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The core action is front-loaded, but the text is bloated and repetitive ('present one approval' and 'ask for one confirmation of the whole displayed plan' restate the same rule, and the expiry/retry_at rule is given twice). Much of it is agent chat-presentation policy rather than guidance for invoking this tool, which crowds out the operative information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a destructive, multi-resource, no-output-schema tool, the description covers failure handling, idempotency, permissions and item-level outcomes. It stops short of describing the receipt/result shape an agent should expect back, and omits the connection_context parameter.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 0% schema description coverage the description must carry the parameters, and it does substantially: approval_ref identifies the stored plan, expected_plan_digest must match, and expected_changes must be the exact values from prepare so the host can inspect targets and effects. Only connection_context is left unexplained.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a concrete verb (execute) and resource (the stored mixed advertising plan identified by approval_ref/expected_plan_digest), and it implicitly distinguishes itself from the prepare step by requiring values 'returned by prepare'. It does not name the sibling confirm/deny/get tools, so the differentiation is inferred rather than stated.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It clearly states the precondition (explicit user approval of every displayed item) and several exclusions (never replay an uncertain write, never retry a host safety denial, changed/expired plans require review), plus the quota-deferred resume path. It never names approval_plan_prepare, approval_plan_deny or approval_plan_get as the alternatives, leaving routing to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All AdsAgent — Google Ads MCP tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?