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 agentmods add skills/cloudx-io/cloudx-sdk-agents/cloudx-auditnpx skills add cloudx-io/cloudx-sdk-agents --skill cloudx-auditgit clone --depth 1 https://github.com/cloudx-io/cloudx-sdk-agentsWrote 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/cloudx-io/cloudx-sdk-agents/cloudx-audit)<a href="https://agentmods.dev/skills/cloudx-io/cloudx-sdk-agents/cloudx-audit"><img src="https://agentmods.dev/badge/skills/cloudx-io/cloudx-sdk-agents/cloudx-audit.svg" alt="Measured on agentmods" 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 | $0.00061 | $0.00634 |
| Opus 5 | $0.00030 | $0.00317 |
| Sonnet 5 | $0.00012 | $0.00127 |
| Haiku 4.5 | $0.00006 | $0.00063 |
Grade A, and why
cloudx-audit 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 4d 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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CloudX Integration Audit
Audit the CloudX integration already present in this app. Do not add new
integration work — for that, use cloudx-integrate.
Iron rule: never judge the integration against APIs from memory. Fetch the
current docs first (https://docs.cloudx.io/llms.txt, then the platform pages —
see ../cloudx-integrate/references/doc-map.md) and diff the code against what
the docs say today. An integration written against an older SDK may compile
and still be wrong or incomplete; the platform changelog page shows what changed.
Steps
- Detect — platform, CloudX SDK dependency and adapters present, ad formats
used, existing mediation SDKs, CMP tooling. Match playbook signals in
../cloudx-integrate/references/playbook-index.mdand read the playbooks that apply. - Fetch — the platform integration overview, the ad-format page for each format the app uses, the platform changelog, and adapter pages for adapters present.
- Diff code against current docs:
- Dependency coordinates and adapter set vs the current documented ones.
For each installed adapter, verify its overview page's requirements are
fully met (extra Maven repositories, manifest/Info.plist entries, native
view requirements). If the publisher's enabled-network set is knowable
(ask, or CloudX MCP/CLI
config/show), flag mismatches both ways: compiled-in adapters for networks not enabled (dead weight) and enabled networks with no adapter in the binary (silent zero contribution). - Initialization: documented location, once, before ad loads, correct config.
- Each ad format: creation, listeners, readiness checks, show requirements, lifecycle/destroy handling per the current docs page.
- Removed/renamed APIs still referenced (changelog is the evidence).
- Dependency coordinates and adapter set vs the current documented ones.
For each installed adapter, verify its overview page's requirements are
fully met (extra Maven repositories, manifest/Info.plist entries, native
view requirements). If the publisher's enabled-network set is knowable
(ask, or CloudX MCP/CLI
- Coexistence — if another mediation stack is present: both init paths
intact, fallback or first-look wiring per the platform's first-look docs
page, no circular loading, per
coexist-mediation.md. - Privacy — CMP writes IAB consent strings before ads load; no usage of
privacy APIs the changelog marks removed; consent forwarded to any
coexisting mediation SDK (per
consent-and-cmp.md). - Report — findings ranked by severity: breaks now / breaks on upgrade / revenue risk / hygiene. Each finding cites the file:line and the docs page that contradicts it. End with the SDK-version gap (dependency present vs current documented version) and the changelog entries in between.
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.
- 4d ago First seen · 48 lines · 61 tokens per session scan A a04c62749369
cloudx-audit is a skill published in the GitHub repository cloudx-io/cloudx-sdk-agents (15 stars, last pushed 1mo ago), licensed MIT. It adds 61 tokens to every session and 634 once invoked, about $0.0003 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.
Other skills, from other repositories
maestro-testing
Generate and manage Maestro test flows for mobile (Android, iOS) and web apps. Use this skill whenever the user mentions mobile testing, UI testing, end-to-end testing, Maestro flows, app automation, test generation from PRDs/specs/user stories, or wants to create/edit/debug YAML test flows. Trigger for any app…
update-store
ストア情報の更新自動化 — スクリーンショット撮影(シミュレーター × モック画面 × Marionette MCP)とメタデータテキスト更新。ストア更新、スクショ更新、App Store / Google Play のメタデータ更新、リリースノート作成の際に使用すること。.
mobile-automation
MCP (dart-mcp + Marionette) を使ったFlutterアプリのE2E自動化・UI検証ガイド。シミュレーターでのUI動作確認、モックプレビュー検証、Bridge経由のE2Eテスト、スクリーンショット撮影など、アプリの動作検証が必要なときに使う。「動作確認して」「UIを検証して」「E2Eテスト」「シミュレーターで確認」「モックで確認」と言われたときや、UI変更後の検証フェーズで使用すること。.
submit-store-review
App Store Connect・Google Playの現在の公開版をAPIで調べ、既存の最新安定ビルドとの差分から4言語のリリースノートを作成し、CLI/APIで審査へ提出・再提出する。新しいビルドを作るrelease-appとは分離し、「最新を審査提出して」「ストア審査へ提出して」「App Reviewに出して」「Google Playの審査を進めて」「却下対応して再提出して」と依頼されたときに使用する。.
triage
Skill "triage" from K9i-0/ccpocket, covering issue / pr triage, 原則, phase 0: 種別判定, prフロー and phase 1: intake / ready判定.
flutter-ui-design
SSOT (Single Source of Truth) + UDF (Unidirectional Data Flow) に基づく設計。.