Skip to main content
Glama

AdsAgent — TikTok Ads MCP

by md.adsagent
OverviewDiscussionsAdmin

List saved creatives

creatives_list
Read-only

List creatives owned by the authenticated tenant with cursor pagination. This is the TikTok AdsAgent tenant library (tiktok.adsagent.md Creatives), not Meta adsagent.md and not TikTok Ads Manager Material Center. READY_LOCAL dashboard rows belong here and are create sources via creative.creative_id; they do not need a TikTok image_id/video_id until after confirm. For images, use one creative_id for one image or ordered creative.image_creative_ids (2..35 JPEG/PNG images) for one carousel; every member must be an eligible image row. ad_params.creative_list contains separate variants, not the ordered cards of one carousel. Rows omit content hashes and storage keys. The legacy download_url is retained as an authenticated same-origin endpoint only after storage verification; unverified historical ready rows report verification_pending. Inspect coverage (create_eligible_count / empty_reason / next_action) plus readiness.create_eligible before using creative_id in campaigns_quick_create. An empty page with coverage.empty_reason=no_tenant_creatives is the only true empty-library signal (unfiltered scope=library plus a tenant-wide existence check). no_creatives_in_query means the current folder/search/media_type/time_filter matched nothing; no_creatives_in_root means scope=root was empty while the tenant owns creatives elsewhere — retry scope=library (default) or call creatives_list_folders. Trust coverage.scope.kind; never describe a root/folder page count as the whole library. Do not treat those as a different tenant.

REQUIRED: none. Optional: scope (library|root, default library — entire tenant library across root and folders; use root for root-only reads), folder_id (str — UUID of one folder), cursor (opaque and scope-bound; reuse it only with the identical query), limit (int 1..50, default 20), search (filename substring, max 200 chars), media_type (all/image/video), time_filter (all/24h/7d/30d), include_incomplete (bool, default true). Returns count, complete, meta.has_more, next_cursor and snapshot_at (legacy top-level has_more remains). Each row exposes creative_id as an alias of id for direct creative.creative_id selection. These describe library pagination, not TikTok upload completion; continue serially with next_cursor instead of client fan-out. EXAMPLE: creatives_list({"search": "hero", "media_type": "video", "limit": 20})

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
limitNo
scopeNolibrary
cursorNo
searchNo
folder_idNo
media_typeNoall
time_filterNoall
connection_contextNo
include_incompleteNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / connection_context
      Added value: +{
+  "pattern": "^mctx_v1_[0-9a-f]{64}$",
+  "type": "string"
+}
  2. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Annotations already declare a safe read (readOnlyHint=true, destructiveHint=false), but the description adds substantial behavior beyond them: cursor is opaque and scope-bound, pagination must continue serially, rows omit content hashes/storage keys, legacy download_url only works after storage verification, and unverified rows report verification_pending. It also documents the coverage/readiness fields that gate downstream use.

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

Conciseness3/5

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

The purpose and scope constraints are front-loaded, and most sentences carry real operational content (empty_reason taxonomy, creative_id aliasing, coverage fields). However, the definition is a dense wall of run-on sentences mixing pagination, folder semantics, carousel rules and empty-state taxonomy, making it harder to scan than it needs to be.

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?

With no output schema, the description must explain returns and does: count, complete, meta.has_more, next_cursor, snapshot_at, the legacy top-level has_more, per-row creative_id as an alias of id, and the coverage fields. Combined with the empty_reason and readiness guidance, an agent has everything needed to call it correctly.

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

Parameters5/5

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

Schema description coverage is 0% and there are no enums in the schema, so the description carries the full burden and does so: scope (library|root with semantics), folder_id, cursor (opaque, scope-bound, reusable only with the identical query), limit (1..50, default 20), search (filename substring, max 200 chars), media_type and time_filter value sets, and include_incomplete. Only connection_context is left undocumented, which is minor.

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 opening sentence gives a specific verb and resource with scope: 'List creatives owned by the authenticated tenant with cursor pagination.' It then explicitly disambiguates the resource from siblings/adjacent systems ('the TikTok AdsAgent tenant library ... not Meta adsagent.md and not TikTok Ads Manager Material Center'), letting an agent separate it from creatives_list_cloud, creatives_list_folders and creatives_get.

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

Usage Guidelines5/5

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

It is unusually explicit about when to use which scope ('library ... entire tenant library ... use root for root-only reads'), when to retry ('no_creatives_in_root ... retry scope=library (default) or call creatives_list_folders'), and where the output is consumed ('before using creative_id in campaigns_quick_create'). It also states the only true empty-library signal, so the agent knows how to distinguish a real empty result from a filter miss.

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

All AdsAgent — TikTok Ads MCP toolsInspect
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