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.
npx skills add Paydirt-AI/paydirt-mcp --skill install-paydirtgit clone --depth 1 https://github.com/Paydirt-AI/paydirt-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/paydirt-ai/paydirt-mcp/install-paydirt)<a href="https://agentmods.dev/skills/paydirt-ai/paydirt-mcp/install-paydirt"><img src="https://agentmods.dev/badge/skills/paydirt-ai/paydirt-mcp/install-paydirt/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/paydirt-ai/paydirt-mcp/install-paydirt"><img src="https://agentmods.dev/badge/skills/paydirt-ai/paydirt-mcp/install-paydirt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00074 | $0.01722 |
| Opus 5.5 | $0.00030 | $0.00689 |
| Sonnet 5.5 | $0.00015 | $0.00344 |
| Haiku 4.5 | $0.00007 | $0.00172 |
Grade A, and why
install-paydirt 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 20d 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.
How it starts
The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Add Paydirt to an iOS app
The human signs in once in the browser. Everything else happens here, in chat. Never ask the human for API keys, tokens, or credentials. You write the SDK key into the project yourself.
1. Inspect, then propose in ONE message
Inspect the project before calling any Paydirt tool: the app target, bundle ID, dependency manager, app entry point, subscription source of truth (RevenueCat, StoreKit, Superwall paywalls, custom), product IDs, and the file where the real cancel flow lives. Note every existing feedback screen and purchase flow; you will preserve all of them.
Then send one message and wait for the reply. Create nothing before the human confirms.
With subscriptions and a cancel flow:
I found your StoreKit subscriptions and the real cancel flow in
Settings/ManageSubscriptionView.swift. I recommend three starting forms: Subscription Cancellation, Trial Cancellation, and Suggest a Feature (in Settings). Responses go to Slack and here in Claude Code. Reply "yes" to set this up, or tell me Slack only, here only, or which form to drop. Nothing is created until you confirm.
Without subscriptions or a cancel flow:
I didn't find subscriptions or a cancel flow, so cancellation feedback has nothing to trigger it yet. I recommend Suggest a Feature (in Settings). Responses go to Slack and here in Claude Code. Reply "yes" to set this up, or tell me Slack only or here only. Nothing is created until you confirm.
Use the real provider, file, and agent name. A plain "yes" accepts the recommended forms and the destinations ["slack", "claude-code"] (use "codex" in Codex). The human may pick Slack only, here only, or drop any form. paydirt_begin_setup without form_plan_confirmed returns this message for you if you need it.
2. Create the app and forms, save delivery, start Slack
Right after the human confirms:
- Call
paydirt_begin_setupwithform_plan_confirmed: true, the confirmeduse_cases,destinations,app_name,bundle_id,subscription_provider,subscription_product_ids,feature_placement, andcancel_flow_file.- Hosted connection (
https://mcp.paydirt.ai/mcp): this creates or reuses the app and forms right away and returns the installation contract. - Local stdio server: it returns
authorization_url. Give it to the human; that page only signs in. Then callpaydirt_finish_setupwithfinish_arguments. Do not poll or sleep while they sign in.
- Hosted connection (
destinationssaves delivery. If you did not pass them, callpaydirt_set_delivery. Setup returnsdelivery_selection_required, never a silent default, when no destination is saved. In that case ask itsquestionexactly.- If Slack was chosen and the contract says it is not connected, call
paydirt_connect_slacknow and share the link: "While I build, connect Slack: ". Do not wait for Slack.
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.
- 20d ago Changed 84fd27955df3
- 23d ago Changed · +48 lines · +28 tokens per session c05a4b91726a
- 2mo ago Changed · +2 lines · +1 tokens per session 51b622b2da1e
- 2mo ago First seen · 22 lines · 45 tokens per session scan A f45ab4cd5fd1
install-paydirt is a skill published in the GitHub repository Paydirt-AI/paydirt-mcp (0 stars, last pushed 21d ago), licensed MIT. It adds 74 tokens to every session and 1,722 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-31.
install-paydirt compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| install-paydirtPaydirt-AI/paydirt-mcp | Skill | 0 | 21d ago | MIT | A | 74 | Claude Code, Codex |
| analyzing-ios-app-security-with-objectionmukul975/Anthropic-Cybersecurity-Skills | Skill | 36k | 2mo ago | Apache-2.0 | A | 49 | Claude Code |
| ios-simulator-skillconorluddy/ios-simulator-skill | Skill | 1.3k | yesterday | MIT | A | 106 | Claude Code, Codex |
| argent-tv-interactsoftware-mansion/argent | Skill | 3.2k | yesterday | Apache-2.0 | A | 107 | Claude Code, Codex |
| mastgChiChou/grapefruit | Skill | 1.4k | 10d ago | MIT | A | 53 | Claude Code, Codex |
| figma-a11y-annotationscvs-health/ios-swiftui-accessibility-techniques | Skill | 388 | yesterday | Apache-2.0 | A | 110 | Claude Code, Codex |
| compose-multiplatformericrisco/rsc-harness | Skill | 503 | 2d ago | MIT | A | 80 | Claude Code, Codex |