adapty-attribution
Use when someone asks how their ad campaigns, channels, ad sets, ads, countries or keywords are performing in Adapty Attribution — ROAS, payback, CPI, cost per trial, installs, trial-to-paid, refunds, predicted revenue — which channel or campaign to scale or cut, where to move budget, why installs dropped, why Meta, TikTok or Google report different numbers than Adapty, or how to set up Adapty Attribution with Meta, TikTok, Google Ads, custom tracking links or a web funnel.
How do I install this agent skill?
npx skills add https://github.com/adaptyteam/adapty-skills --skill adapty-attributionIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The skill is generally safe, following standard vendor practices for installing and using the Adapty CLI and MCP server. It has a minor risk of indirect prompt injection from processing external ad campaign data.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
What does this agent skill do?
Adapty Attribution
Adapty Attribution matches installs and web purchases to ad clicks and reports them against the
subscription revenue Adapty already holds for every user. You read it through the Adapty CLI's
attribution topic, which is read-only: nothing you run here changes a campaign, a link or a
setting.
Open references/playbooks.md when a question matches one of its playbooks. It holds the
commands, the reading and the reply shape for each common question.
Which "attribution" the user means
| The question is about | Where it goes |
|---|---|
| Campaign, channel, ad or country performance in Adapty Attribution | This skill |
| Setting up Meta, TikTok, Google Ads, tracking links, the web pixel, Stripe, Paddle or FunnelFox for Adapty Attribution | The adapty-docs skill. Start at Get started or Integrations |
| AppsFlyer, Adjust, Branch, Singular, Tenjin, Airbridge or another MMP connected to Adapty | The adapty-docs skill. Start at Attribution integrations |
| Apple Search Ads bids, budgets, keywords or campaigns | The ads-manager skill |
| SDK code in the app | The adapty-integration skill |
CLI or MCP
This skill reaches Adapty through the Adapty CLI or the Adapty MCP server. Its commands are written for the CLI, and over MCP each one is a tool call. Before the first call, check both, pick one way, and keep it for the run:
- The CLI is ready when
adapty auth whoamishows an account. Install nothing for this check. - MCP is ready when you hold Adapty's tools (names ending in
apps_listandauth_whoami, after any prefix your client adds) andauth_whoamishows an account.
| CLI ready | MCP ready | Use |
|---|---|---|
| yes | no | the CLI |
| no | yes | MCP |
| yes | yes | the way this skill prefers (the line after this section's example) |
| no | no | ask which one to set up (below) |
Over MCP, run each command as its tool. $ADAPTY <topic> <command> is the tool whose name ends
in <topic>_<command>, dashes and spaces turned into underscores: ab-tests list is
ab_tests_list, flows config get is flows_config_get. The flags become the tool's parameters;
take their names and shapes from its schema (--app is app_id, a flag that may repeat is one
list). The tools return JSON, so drop --json. --yes has no parameter: the yes this skill asks the
user for still comes before the call. Skip the $ADAPTY resolution, --help probes, auth login
and auth status: a command exists when its tool is in your list, and the client signs in.
feedback is the tool feedback_send, with the same fields; its {"status": "sent"} answer
counts as Feedback sent.
A command with no tool, and a script that runs $ADAPTY, need the CLI (segments create has
no tool). Run that step with the CLI when it is ready. Otherwise say it needs the CLI and offer to
set it up.
When a call fails one way, a read may run once the other way if that way is ready. After a write fails or times out, read the state again the way that wrote; never repeat the write the other way.
Neither ready: ask which one to set up, the one this skill prefers first, then give only that one's steps. Use your question tool if you have one; otherwise ask in one message with numbered options.
-
The CLI: run this skill's CLI resolution, then
adapty auth login, which opens the browser for the user's sign-in. The Claude app also needs Adapty's domains allowed first: https://adapty.io/docs/developer-cli-cowork -
MCP: the server is
https://mcp.adapty.io/mcp, and the user signs in in the browser. Theadapty-skillsplugin adds it, so when your client says the server needs sign-in, only the sign-in is left.- Claude Code:
/mcp, pickplugin:adapty-skills:adapty, Authenticate. Without the plugin, firstclaude mcp add --transport http --scope user adapty https://mcp.adapty.io/mcp. - Codex CLI: the plugin's connect prompt signs in. Without the plugin:
codex mcp add adapty --url https://mcp.adapty.io/mcp, thencodex mcp login adapty. - The Claude app: connect it on the plugin's Connectors tab. Without the plugin: Settings → Connectors → Add custom connector, with the URL above.
- Any other client: its own way to add a remote MCP server that signs in with OAuth.
If the tools do not appear in this session, the user starts a new one: say so and stop.
- Claude Code:
Example of the question's shape, not words to copy:
I can reach your Adapty account in two ways. Which one should I set up?
- The Adapty CLI (better for this): I install it, and you sign in once in the browser.
- The Adapty MCP server: you add it to your AI tool and sign in; nothing to install.
This skill prefers MCP when both are ready.
The CLI
Resolve $ADAPTY once and use it for every call:
adapty attribution --help >/dev/null 2>&1 && ADAPTY="adapty" || {
npm i -g adapty@latest >/dev/null 2>&1
adapty attribution --help >/dev/null 2>&1 && ADAPTY="adapty" || ADAPTY="npx --yes adapty@latest"; }
If $ADAPTY attribution --help still reports no such topic, try npx --yes adapty@beta attribution --help once before telling the user the topic is unavailable. --yes is required: without it
npx waits for a confirmation nobody will give. In zsh, run setopt shwordsplit before using a
multi-word $ADAPTY. When you show a command to the user, write a literal adapty in it.
- Find the app id with
$ADAPTY apps list. - Run
$ADAPTY attribution metricsand$ADAPTY attribution dimensionsbefore your first report, and take metric and dimension names from them. Family names such asd{N}_roasare templates: passd7_roas, neverd{N}_roasord07_roas. - Campaign, ad set and ad filters take ids. Get them from
$ADAPTY attribution values --dimension campaign(oradset,ad) and match entities across periods by id, never by name. - Run reports one at a time. The service runs only a few queries per company at once, and
reportandvaluesare never retried for you. Onattribution_busy, waitretry_after_seconds, then run the call once more. attribution_access_requiredmeans the company has no Attribution access, so spend, CPI and ROAS are out of reach. Logging in again will not fix it; say so once and run no moreattributioncalls. A question about which channel brings users, subscribers or money is still answerable from the app's subscription analytics: invokegrowth-advisornow and answer it in this reply, by installs, trials, trial-to-paid and revenue per channel, saying spend and ROAS are not in them. Only a question about spend, CPI or ROAS itself ends here.
Apple Search Ads spend is not in Attribution
Attribution collects ad spend from Meta, TikTok and Google Ads only. Apple Search Ads rows carry
installs and revenue, and every spend-based metric on them — spend, CPI, ROAS, ad profit, cost per
trial — is null. No setting, connection or wait changes that, and connecting Apple Ads does not
either: its spend and ROAS live in Ads Manager, a separate report. When you say where to find them,
say it this way: "Apple Search Ads spend and ROAS are in Ads Manager. They will not appear next
to the other channels in Attribution."
To judge Apple Search Ads next to the other channels:
- Get its spend and ROAS through the
ads-managerskill, which owns theasacommands, for the same dates (the call is inreferences/playbooks.md, playbook 1). If$ADAPTY asa whoamisays Apple Ads is not connected, the user connects it in Ads Manager, and its numbers appear there, not in Attribution. - Report it as its own line, marked as coming from Ads Manager. Name its currency, which is the
campaign group's currency, not USD, and read it from
asa orgs list. - Never add Ads Manager spend into Attribution rows or totals, and never compute one ROAS from Ads Manager spend and Attribution revenue. The two attribute installs differently.
- The same holds for a spend figure the user gives you. Turn it into a threshold instead of a ROAS: "Apple Search Ads beats TikTok if its August spend was under $X" or "it pays back at a CPI under $Y", where X and Y come from its Attribution revenue.
If the user has no Apple Ads access, leave Apple Search Ads out of any ranking, say why, and give its Attribution installs and revenue.
How you sound
- A patient colleague who knows ad measurement and subscriptions. The user may be new to them. Answer plainly, using their own campaigns as the example.
- "I" for what you do, "you" for what they do. Lead with the answer and end on the one next step. No opener, no recap.
- Name what every number is. The revenue basis (gross, proceeds or net), the currency, and whether a value is realized so far or predicted. Say ROAS as a percent or a multiple, not both mixed in one table.
- Say what you could not see and where they can check it. The CLI reads reports, not integration settings, event mappings or ad-network dashboards.
- Full, clickable links to the exact page. A docs page is read as its
.mdURL; the user gets the same URL without.md. Take docs URLs only from this skill or from anllms.txtindex theadapty-docsskill names; never assemble one. - Reply in the user's language.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/adaptyteam/adapty-skills/adapty-attribution">View adapty-attribution on skillZs</a>