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 mgcrea/mcp-appstore-connect --skill appstore-release-prepgit clone --depth 1 https://github.com/mgcrea/mcp-appstore-connectWrote 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/mgcrea/mcp-appstore-connect/appstore-release-prep)<a href="https://agentmods.dev/skills/mgcrea/mcp-appstore-connect/appstore-release-prep"><img src="https://agentmods.dev/badge/skills/mgcrea/mcp-appstore-connect/appstore-release-prep/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/mgcrea/mcp-appstore-connect/appstore-release-prep"><img src="https://agentmods.dev/badge/skills/mgcrea/mcp-appstore-connect/appstore-release-prep.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 | $0.00277 | $0.07874 |
| Opus 5.5 | $0.00111 | $0.03150 |
| Sonnet 5.5 | $0.00055 | $0.01575 |
| Haiku 4.5 | $0.00028 | $0.00787 |
Grade A, and why
appstore-release-prep 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 3d 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 — 557 lines — stays where its author put it; the contents beside it link to each section on GitHub.
App Store release prep
Turn "what changed since the last release" into two documents that stay in sync: a CHANGELOG entry (for developers, complete) and the App Store Connect fields (for customers, ruthlessly short and hard-limited).
The failure modes here are boring and specific, and they are why this skill exists: people guess the version, miss features that shipped without ever being written down, and blow a 4,000-character limit they never measured. Run the audit first — it answers all three deterministically — then spend your judgment on the copy.
1. Audit before writing anything
python3 ${CLAUDE_SKILL_DIR}/scripts/audit_release.py --repo <repo> # add --json for machine output
It reports: the shipping version and the last documented one, the release boundary commit and everything since it, commits that shipped in the last release but were never announced, every store field measured against Apple's limit, em dashes in prose, screenshot-config drift, and the captions of any cross-platform family images (listed for you to check, never judged). It exits non-zero if a field is over or missing, so it can gate a release.
The source is auto-detected: a metadata tree wins, else APPSTORE.md, STORES.md
and friends. Trees are found from every .listing.json, plus fastlane/metadata/ or
Listing/ at --repo and beside the Xcode project, and a root holding platform folders
(Listing/macos/ + Listing/ios/) is one tree per platform. Every tree is audited
and gates the exit code, each labelled with its platform. --repo can be a monorepo
root: the app directory is found from its .xcodeproj, its CHANGELOG is looked for there
first, and commits that touch only a sibling app (apps/website) are left out of the news.
Pass --metadata-root <path> for a tree that was moved and has no sidecar, or to audit
one tree only. Pass --fields-file <path>
to force a document, or --locale to audit a locale other than the primary one.
Only a dated heading is a release. ## [1.0.0] - Unreleased is the entry being written:
the audit says so, and since nothing has shipped it lists the commits made since the
CHANGELOG was last edited, which are the ones that entry cannot mention yet.
Check the source it reports — the header says which file or directory the numbers
came from.
If it reports "file does not exist" for a project that plainly has store copy, you
pointed it at the wrong place; do not take that as license to write every field from
scratch, or you will replace live copy with a rewrite nobody asked for.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago Changed · +9 lines 87c6ba8eb6d8
- 1mo ago Changed · +45 lines b3301b7005fb
- 1mo ago Changed 8853799432cd
- 1mo ago Changed · +38 lines 83002273e98b
- 1mo ago Changed · +2 lines 0b7bbfcef89e
- 2mo ago Changed · +55 lines 5b8840413966
- 2mo ago Changed · +13 lines 1984abd5b44b
- 2mo ago First seen · 395 lines · 277 tokens per session scan A d0ca386259aa
appstore-release-prep is a skill published in the GitHub repository mgcrea/mcp-appstore-connect (2 stars, last pushed yesterday), licensed MIT. It adds 277 tokens to every session and 7,874 once invoked, about $0.0011 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.
appstore-release-prep compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| appstore-release-prepmgcrea/mcp-appstore-connect | Skill | 2 | yesterday | MIT | A | 277 | Claude Code |
| orca-emulator-androidstablyai/orca | Skill | 117k | yesterday | MIT | A | 102 | Claude Code, Codex |
| glance-canonical-layoutsandroid/skills | Skill | 10k | yesterday | Apache-2.0 | A | 56 | Claude Code |
| android-tombstone-symbolicationdotnet/skills | Skill | 5.8k | yesterday | MIT | A | 176 | Claude Code |
| dogfoodcallstack/agent-device | Skill | 5.3k | yesterday | MIT | A | 55 | Claude Code |
| winapp-mauimicrosoft/winappCli | Skill | 1.4k | yesterday | MIT | A | 68 | Claude Code |
| app_remixrome-os/rome | Skill | 889 | yesterday | MIT | A | 50 | Claude Code, Codex |