BloomDesktop: Skill for Claude Code

.claude/skills/merge-crowdin-translations/SKILL.md

merge-crowdin-translations is a skill for Claude Code from BloomBooks/BloomDesktop. It costs 96 tokens per session (3,999 once invoked), scanned A, original, MIT.

A procedure for bringing translations collected by Crowdin into the main code branch and selected release branches. Crowdin is a translation service that stores translators' changes in a pull request.

In plain words
What is it for?
Use it to merge Crowdin translations, update `master`, and copy the result to chosen release branches with Git.
Why use it?
It turns many translation commits into a controlled update and lets you check what each branch gains or loses.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is BloomBooks/BloomDesktop's own configuration. It tells Claude Code how to work on BloomDesktop itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything BloomDesktop configures →

Reuse

Borrowing it

Nothing to install: this file belongs to BloomBooks/BloomDesktop. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/BloomBooks/BloomDesktop/master/.claude/skills/merge-crowdin-translations/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/BloomBooks/BloomDesktop

Made for: Claude Code.

Wrote 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.

agentmods badge for merge-crowdin-translations

README.md
[![agentmods](https://agentmods.dev/badge/skills/bloombooks/bloomdesktop/merge-crowdin-translations/github.svg)](https://agentmods.dev/skills/bloombooks/bloomdesktop/merge-crowdin-translations)
Your own site
<a href="https://agentmods.dev/skills/bloombooks/bloomdesktop/merge-crowdin-translations"><img src="https://agentmods.dev/badge/skills/bloombooks/bloomdesktop/merge-crowdin-translations/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.

agentmods 80×15 button for merge-crowdin-translations

Your own site · 80×15
<a href="https://agentmods.dev/skills/bloombooks/bloomdesktop/merge-crowdin-translations"><img src="https://agentmods.dev/badge/skills/bloombooks/bloomdesktop/merge-crowdin-translations.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 96 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,999 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00096 $0.03999
Opus 5.5 $0.00038 $0.01600
Sonnet 5.5 $0.00019 $0.00800
Haiku 4.5 $0.00010 $0.00400

Measured yesterday against content hash 918bd56853dc, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

merge-crowdin-translations scanned grade A with 1 finding 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 yesterday.

The scan reads SKILL.md. This mod also ships 1 executable file (audit_translations.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -sL -o pkg.zip https://api.nuget.org/v3-flatcontainer/l10nsharp.checkorfixxliff/<ver>/l10nsharp.checkorfixxliff.<ver>.nupkg
.claude/skills/merge-crowdin-translations/SKILL.md · 275 lines

How it starts

The opening of the file, as written. The whole thing — 275 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Merge Crowdin translations

Crowdin keeps one pull request open from l10n_master into master and adds a commit to it every ten minutes while translators work, so it holds hundreds or thousands of "New translations" commits. Every so often we bring in what translators have done since the last time: squash those commits into one, commit it directly on master, and cherry-pick it directly onto release branches. No pull requests are involved; the developer approves each push instead. The agent does all of it except pausing and resuming Crowdin's sync, which the developer does in the Crowdin web page.

How Crowdin syncs, what edits to xliff files do to translations, and why translations travel master-first are in DistFiles/localization/README.md. Read it if anything here looks odd.

Doing this by hand: every step below is a plain git, gh or python command; run them yourself in order. Where the agent asks, you decide.

Asking the developer

Every question goes to the developer through the decider skill, never as a terminal question or a list in chat. Prepare everything that can be prepared first (steps 2–5), so that one form carries all the decisions at once (step 6). Ask again with a new form only when something forces it: a conflict in step 3, a problem that stops the run, or a push rejected because a branch moved.

Pausing and resuming the sync (steps 1 and 9) are not decisions; ask for those in a plain message.

Nothing is pushed without the developer's answer on the form. Each commit is made locally on top of its target branch, and pushing it is the merge.

Where to work

Work in a clean worktree where you can switch branches freely, not one holding someone's work. If you need one, make it from master (git worktree add <scratch>/crowdin-work origin/master) and remove it at the end. On Windows, if git worktree remove fails with "Permission denied", delete the folder and run git worktree prune. Run this skill's script from wherever the skill itself is.

Read the full file on GitHub · 275 lines

Files

What ships with it

1 file 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.

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. yesterday First seen · 275 lines · 96 tokens per session scan A 918bd56853dc

Subscribe to this mod's changes

merge-crowdin-translations is a skill published in the GitHub repository BloomBooks/BloomDesktop (43 stars, last pushed today), licensed MIT. It adds 96 tokens to every session and 3,999 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-29.