Taseli

Propose an account change

propose_change

Prepares one change to a running ad account as a proposal the owner approves in Taseli: pause or resume a campaign (or a Meta ad set, a Google ad group), a new daily budget, a target (Google target CPA or ROAS, a Meta or TikTok cost cap or bid cap, a Meta ROAS floor; a proposal from outside moves it at most 15% from the current one), or blocked searches (Google Search negatives, for one campaign or every running search campaign). Nothing changes in the account here, and it never runs by itself: the owner approves it in Taseli, which then runs it, reads it back and watches it 7 days. The reason goes in why, and the owner reads it on the card. A proposal from outside follows Taseli's rules and may be held (declined lately, a campaign still learning, the owner's rules); owner_asked with the owner's words in quote records that the owner asked for exactly this. restructure (Meta and TikTok): Taseli's structure check builds the change from the account's numbers (which ad sets or ad groups merge, which ads move, what budget; total spend stays the same) and says why when none is due. Not for new campaigns or for new creatives in running campaigns. The same change again returns the same proposal.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "change": "<change>",
      "why": "<why>"
    },
    "name": "propose_change"
  }
}

Input Schema

Name	Required	Description
`fix`	No	restructure: merge like ad sets / ad groups into one (Google: thin ad groups of one theme, or duplicate keywords kept once), a closer goal (Meta / TikTok: checkout or add to cart instead of a rare purchase), one campaign budget (Meta), own_goal (Google: a campaign counts only the real results), or brand (Google: Performance Max kept off brand searches). Leave out for the most valuable one.
`why`	Yes	One short line the owner reads on the card, in the conversation's language: why this change, from this account's facts.
`quote`	No	With owner_asked: the owner's own words, as written.
`terms`	No	block_searches: the search words to block (phrase match).
`change`	Yes
`account`	No	An ad account's name or id; needed when the business has more than one.
`campaign`	No	The campaign (or ad set / ad group) name exactly as Taseli shows it, or its id. For block_searches, omit for every running search campaign.
`language`	No	Answer language. The business's language by default.
`owner_asked`	No	true only when the owner asked for exactly this change in this conversation.
`target_kind`	No	target: cpa = target cost per result (Google) or cost cap (Meta, TikTok); roas = target return (Google) or ROAS floor (Meta); bid_cap = Meta bid cap or TikTok bid.
`daily_budget`	No	budget: the new daily budget, in the account's currency.
`target_value`	No	target: the new value. Money in the account's currency; roas as a ratio (3.5 = 350%).
`budget_change_pct`	No	budget, instead of daily_budget: a percent step from the current daily budget, e.g. 15 or -20; Taseli reads the live budget and works out the amount.
`target_change_pct`	No	target, instead of target_value: a percent step from the current target, e.g. 10 or -10; Taseli reads the current one.

Output Schema

Name	Required	Description
`error`	No	not_found, ambiguous, read_only, role, invalid, declined_lately, skipped or failed.
`detail`	No
`status`	No	waiting: on the owner's approval screen.
`account`	No
`candidates`	No
`proposalId`	No
`approvalUrl`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedNov 6, 2026

addedInput schema / properties / budget_change_pct

Added value: +{
+  "description": "budget, instead of daily_budget: a percent step from the current daily budget, e.g. 15 or -20; Taseli reads the live budget and works out the amount.",
+  "type": "number"
+}

First observedNov 3, 2026

TDQS

A4.6/5.0

Behavior5/5

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

Beyond the annotations it discloses the key behaviors an agent needs: nothing is executed here, the owner approves in Taseli which then runs it, reads it back and watches it for 7 days, outside proposals are capped at 15% moves on targets and may be held by Taseli's rules. It also discloses the repeat-call behavior ('the same change again returns the same proposal'), which is useful context even though it sits in mild tension with idempotentHint=false.

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

Conciseness4/5

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

Front-loaded with the core statement — what it prepares, that nothing executes, and who approves — before enumerating change types. It is dense and heavily parenthetical, and the target/restructure details could be compressed, but nearly every clause carries decision-relevant information.

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

Completeness5/5

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

For a 14-parameter tool, the description covers the full decision surface: what changes, what is prohibited, what happens after approval, when a proposal is held, and how to record owner-initiated requests. With an output schema present, it appropriately does not spend words on return values.

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?

Schema coverage is already 93%, so the baseline is 3, and the description still adds real meaning: it maps `change` values to platform-specific mechanics, explains the `fix` values for restructure and its default, clarifies that `campaign` is omitted for block_searches to hit every running search campaign, and defines `target_kind` per platform. Some semantics (e.g. how `terms` phrase matching behaves, currency edge cases) remain solely in the schema.

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

Purpose5/5

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

The description states a specific verb and resource — 'Prepares one change to a running ad account as a proposal the owner approves in Taseli' — and enumerates the exact change types (pause/resume, budget, target, blocked searches, restructure) with per-platform scope. It also draws a clear boundary against the neighboring propose_campaign case: 'Not for new campaigns or for new creatives in running campaigns.'

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?

Explicit when-not guidance is present ('Not for new campaigns…', 'Leave out for the most valuable one' for `fix`), plus context on when a proposal may be held (declined lately, learning campaign, owner's rules) and when to set `owner_asked`/`quote`. It stops short of naming the sibling tool an agent should switch to for new campaigns or new creatives, so routing is implied rather than spelled out.

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

All Taseli 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?