launch-pmax

A command workflow for creating a Google Ads Performance Max campaign. Performance Max is a Google advertising campaign type that distributes ads across several Google services using shared creative assets.

In plain words
What is it for?
Use it to create a campaign, asset group, audience signals, and, for online shops, Merchant Center product listings.
Why use it?
It gathers the campaign settings and prepares the campaign paused, so the setup can be reviewed before it starts spending.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/markifact/markifact-mcp/launch-pmax
Clone the repo
git clone --depth 1 https://github.com/markifact/markifact-mcp
Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,468 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00041 $0.01468
Opus 5 $0.00020 $0.00734
Sonnet 5 $0.00008 $0.00294
Haiku 4.5 $0.00004 $0.00147

Measured 2d ago against content hash 189fa837703d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

launch-pmax scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/launch-pmax.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Goal

Stand up a complete PMax campaign with one fully-loaded asset group and (when applicable) Merchant Center listing scope. Always start paused.

Inputs to confirm with the user (batch these)

  1. Account name (substring) — required.
  2. Goal: Sales (ROAS-driven, usually with Merchant Center) or Leads (CPA-driven). Required.
  3. Daily budget. Required.
  4. Bidding: MAXIMIZE_CONVERSIONS (default for leads/no history), MAXIMIZE_CONVERSION_VALUE (default for sales), or with target (TARGET_CPA / TARGET_ROAS). Required.
  5. Geo + language. Required.
  6. Final URL(s). Required.
  7. Brand name + 1-line value prop. Required.
  8. Asset inventory the user can provide: logos (1:1 + 4:1), images (landscape, square, portrait), videos (optional but Google will auto-generate if missing). Ask which they have; you'll generate text assets, but Google won't accept text-only PMax for most goals.
  9. For Sales: Merchant Center ID + product scope (all products, by brand, by product type, by custom label). Required.
  10. Audience signals (optional but recommended): existing customer lists, website visitors, custom segments by interest/intent. These are hints, not targeting.

Workflow

  1. Discover with find_operations. Look for: gads_select_accounts, gads_create_campaign, gads_create_pmax_asset_group, gads_create_image_asset, gads_create_headline_asset, gads_create_description_asset, gads_update_audience_signals, gads_update_listing_groups, gads_pmax_channel_split, gads_search_geo_targets.
  2. Inspect each with get_operation_inputs.
  3. Resolve account: gads_select_accounts with substring match.
  4. Resolve geo: gads_search_geo_targets.
  5. Create campaign (PAUSED). Set advertising_channel_type=PERFORMANCE_MAX, the bidding strategy + target, daily budget, geo + language. For Sales: link the Merchant Center ID.
  6. Create asset group under the campaign → returns asset_group_id.
  7. Upload image assets with gads_create_image_asset for each user-provided image and logo. Capture the returned asset IDs.
  8. Create text assets with gads_create_headline_asset and gads_create_description_asset — generate the variety described in the rules below.
  9. Attach assets to the asset group as the create-asset ops require (see each op's schema — some accept an asset_group_id directly).
  10. Audience signalsgads_update_audience_signals with chosen segments. Skip if user has none, but explain that it slows ramp-up.
  11. Listing groups (Sales only) — gads_update_listing_groups to scope which products PMax can advertise. Default: all products. Refine if the user wants brand/category exclusions.
  12. Confirm: campaign id, asset group id, asset count by type, audience signal count, listing scope. Remind user it's paused.

Read the full file on GitHub · 81 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 81 lines · 41 tokens per session scan A 189fa837703d

Subscribe to this mod's changes

launch-pmax is a command published in the GitHub repository markifact/markifact-mcp (48 stars, last pushed 14d ago), licensed MIT. It adds 41 tokens to every session and 1,468 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.