Agent skill

Brand Region Client

by schmidtwmark in schmidtwmark/BetterBlue

Add a new brand/region client to BetterBlueKit, or fix errors in an existing one.

MITAuto-check passedMobile

Install Brand Region Client

skills CLI
$ npx skills add schmidtwmark/BetterBlue --skill brand-region-client -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install schmidtwmark/BetterBlue brand-region-client --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/schmidtwmark/BetterBlue.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/brand-region-client .claude/skills/brand-region-client && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
brand-region-client
GitHub stars
132
Token cost
~2.1k tokens
SKILL.md length
940 words
Files
3 (incl. references)
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

Add a new brand/region client to BetterBlueKit, or fix errors in an existing one.

  • Works in 6 steps: Confirm scope → Find an authoritative reference → Pick the reference client to copy → …
  • The user asks to add Kia Canada
  • SKILL.md covers Phase 0 — Confirm scope, Phase 1 — Find an…, Phase 2 — Pick the reference… and Phase 3 — Scaffold the client, plus 4 more sections
  • Calls swift; reaches raw.githubusercontent.com

What it does

Brand Region Client is an agent skill from schmidtwmark/BetterBlue. Add a new brand/region client to BetterBlueKit, or fix errors in an existing one. Use when the user asks to "add Kia Canada", "implement Hyundai Australia", "fix the European client", or similar. The skill gathers an API reference (open-source implementation or user-captured proxy logs), writes the client following the repo's four-file layout (+Commands / +Parsing / main / registration), and then iterates by running swift run bbcli ... — fixing each error against the reference until every menu action passes.

Its SKILL.md is about 2.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/bbcli-cheatsheet.md` and `references/reference-implementations.md`).

It sits in Mobile, covering iOS development. The repository describes itself as: An iOS native app for controlling Kia and Hyundai vehicles. The licence is MIT.

When your agent uses it

  • The user asks to add Kia Canada
  • Implement Hyundai Australia
  • Fix the European client

Example prompts

  • “add Kia Canada”
  • “implement Hyundai Australia”
  • “fix the European client”
  • “/brand-region-client”

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Confirm scope
  2. Find an authoritative reference
  3. Pick the reference client to copy
  4. Scaffold the client
  5. Iterate with bbcli
  6. Verify and document

What it can do on your machine

Read from SKILL.md and the folder at commit dc67480. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • swift

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • raw.githubusercontent.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Brand Region Client loads about 2.1k tokens when it runs, and up to ~3.4k if it reads all its reference files. Until then it costs about 134 tokens; SKILL.md has 940 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~134
When it runs · the whole SKILL.md, loaded when a task matches
~2.1k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from schmidtwmark/BetterBlue at commit dc67480, republished under its MIT licence (© schmidtwmark). 940 words, ~2,104 tokens.

Download SKILL.mdSave it as .claude/skills/brand-region-client/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
brand-region-client
description
Add a new brand/region client to BetterBlueKit, or fix errors in an existing one. Use when the user asks to "add Kia Canada", "implement Hyundai Australia", "fix the European client", or similar. The skill gathers an API reference (open-source implementation or user-captured proxy logs), writes the client following the repo's four-file layout (+Commands / +Parsing / main / registration), and then iterates by running `swift run bbcli ...` — fixing each error against the reference until every menu action passes.
tools
Read, Write, Edit, Glob, Grep, Bash, WebFetch, WebSearch, AskUserQuestion

Brand / Region Client Builder

Builds or fixes a BetterBlueKit API client for one (Brand, Region) pair. The hard part isn't Swift, it's figuring out the region's undocumented API — so every step is anchored to either an open-source reference or user-captured proxy logs.

Before starting, read docs/AddingNewRegion.md end-to-end. That guide is the rulebook; this skill is the execution playbook.

Phase 0 — Confirm scope

Ask the user (via AskUserQuestion):

  1. Brand and region — e.g. "Kia Canada".
  2. New client or fix? — greenfield implementation vs. patching an existing one. If patching, ask which bbcli command is failing and with what error.
  3. Test credentials — do they have an account they can log in with? Tests against live accounts are how we validate each step.

Do NOT proceed past this phase without explicit answers. Guessing the brand or region silently rewires the factory.

Phase 1 — Find an authoritative reference

A wrong reference wastes hours. Check in this order:

  1. hyundai_kia_connect_api (Python) — the broadest coverage, especially for non-USA regions. Fetch the region's module and the shared vehicle.py / api_impl.py with WebFetch:
    • https://raw.githubusercontent.com/Hyundai-Kia-Connect/hyundai_kia_connect_api/master/hyundai_kia_connect_api/
  2. bluelinky (TypeScript)
  3. scriptable-egmp (TypeScript)

For each reference, record:

  • Base URL / host
  • Login endpoint + body shape
  • Auth token header name (Authorization vs. accessToken vs. Access-Token — these differ per region)
  • Vehicle list endpoint
  • Vehicle status endpoint (and whether cached/real-time live at the same URL or different URLs)
  • Command endpoints for: lock, unlock, start climate, stop climate, start charge, stop charge

If NO reference covers this region (e.g. Kia Australia), skip to Phase 1b.

Phase 1b — User-captured proxy logs

Ask the user to capture traffic from the official app using Proxyman / mitmproxy / Charles. The AddingNewRegion guide has the step-by-step. Request, at minimum:

  • Login (one request)
  • Vehicle list (one request)
  • Vehicle status (refreshed once so both endpoints hit if they differ)
  • Lock + unlock (one each)
  • Start climate with a known temperature

Have them export to a file and share the path. Then use Read on each to study request/response pairs. Treat every token and cookie as sensitive — never paste raw captures into code or commit them.

Phase 2 — Pick the reference client to copy

Match the new region against an existing BetterBlueKit client with the closest auth model. The right starting point saves the most work:

If the new region uses...Copy the layout of
Header-based access token, no MFAHyundaiUSA
Cloudflare cookie + per-command PINHyundaiCanada
OAuth bearer token, CCS2 payloadsHyundaiEurope
MFA-gated login, separate real-time endpointKiaUSA

Use Read to open the chosen client plus its +Commands.swift and +Parsing.swift extensions. Note the file naming convention — every client follows the same three-file split.

Phase 3 — Scaffold the client

Create three files under BetterBlueKit/Sources/BetterBlueKit/API/<Brand><Region>/:

<Brand><Region>APIClient.swift            // login, fetchVehicles, fetchVehicleStatus, sendCommand
<Brand><Region>APIClient+Commands.swift   // commandURL, commandBody, method helpers
<Brand><Region>APIClient+Parsing.swift    // response → VehicleStatus / Vehicle mappers

The class MUST subclass APIClientBase and conform to APIClientProtocol. Start every method that isn't yet implemented by throwing APIError.regionNotSupported("not yet implemented") — a failing bbcli run is more actionable than a compile error.

Then register in APIClientFactory.swift:

  1. Add a case in the brand-appropriate private createXClient function.
  2. Add the region to supportedRegions(for:) (or betaRegions(for:) if you want a visible "beta" marker).

Phase 4 — Iterate with bbcli

This is the core loop. Run from the package root so Swift Package Manager resolves:

bash
cd BetterBlueKit
swift build 2>&1 | tail -20   # compile first; fix any Swift errors
swift run bbcli -b <brand> -r <region> -u <user> -p <pass> [--pin <pin>]

Inside bbcli, work through the menu in this order:

  1. Fetch Vehicles (menu 1) — validates login + token plumbing. Usually the first thing that breaks.
  2. Fetch Vehicle Status (menu 2) — validates per-vehicle auth and the status parser.
  3. Lock / Unlock (menu 3 / 4) — simplest commands, minimal payloads.
  4. Start / Stop Climate (menu 5 / 6) — largest command bodies; temperature encoding often differs per region.
  5. Start / Stop Charge (menu 7 / 8), Charge Limits (menu 9), EV Trip Details (menu 10).
Show full SKILL.md (343 more words)Show less
When a step fails

For each error, gather evidence from all three sources:

  1. Read the exact error bbcli printed — status code, body. (Re-run with --no-redaction to see raw tokens if the redacted output isn't enough.)
  2. Read the equivalent request/response in the reference implementation or proxy log. Diff the headers, body, and endpoint path against what bbcli just sent.
  3. Edit the client file to match the reference. Common diffs: different Content-Type, different auth header name, a nested wrapper key (result.vehicles vs. payload.vehicles), a missing deviceid, or a different date format.

Rebuild and re-run the same menu item. Advance to the next only when the current one returns parseable data.

Parsing iteration

Response parsers are the most error-prone part. When a response comes back 200 but the parser throws, don't hammer the live API — dump the responseBody to a file and iterate offline:

bash
swift run bbcli parse -b <brand> -r <region> -t vehicleStatus ./fixture.json
swift run bbcli parse -b <brand> -r <region> -t vehicles ./fixture.json

bbcli parse unwraps a top-level "responseBody" automatically — paste a full debug export and it Just Works. Save working fixtures under BetterBlueKit/Tests/BetterBlueKitTests/Fixtures/<Region>/ so regressions are catchable.

Phase 5 — Verify and document

  1. Run every menu item end-to-end against a live account. If any still fails, it's not done.
  2. Update Troubleshooting.md — flip the region row from ❌ to ✅ (or ✅ beta).
  3. If you added to betaRegions(for:) and everything now works, promote to the supported list.
  4. Summarize for the user: what changed, which bbcli menu items passed, any edge cases left as TODOs.

Scope rules

  • Never commit secrets. Proxy captures and debug exports can contain live tokens. Scrub before embedding any example in code comments or fixtures.
  • Don't invent endpoints. Every URL, header, and body field must be traceable to either a reference implementation or a proxy capture the user provided.
  • Don't skip bbcli validation. A client that "looks right" without being exercised by bbcli has a ~0% success rate in production.
  • Ask when stuck. If a reference disagrees with a proxy log, surface the conflict to the user rather than guessing.

References

See references/ for condensed cheat sheets:

© schmidtwmark, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in .claude/skills/brand-region-client of schmidtwmark/BetterBlue.

  • SKILL.md
  • references/bbcli-cheatsheet.md
  • references/reference-implementations.md

Open the folder on GitHubat commit dc67480

Compare with similar skills

Brand Region Client next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Brand Region Client compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Brand Region Client this skillschmidtwmark/BetterBlue132—~2.1kAutomated safety check: PassMIT
Swiftui Protwostraws/SwiftUI-Agent-Skill5.2k2 repos~1.5kAutomated safety check: PassMIT
Swiftui UI PatternsAFK-surf/OpenBridge4304 repos~887Automated safety check: PassMIT
Swiftui Performance Auditharperreed/dotfiles3348 repos~1.4kAutomated safety check: PassNone
Swift Testing ExpertAvdLee/Swift-Testing-Agent-Skill4701 repos~1.2kAutomated safety check: PassMIT
Hig Project Contextraintree-technology/hig-doctor1435 repos~1.2kAutomated safety check: PassMIT

Similar skills

  • Swiftui Pro

    twostraws/SwiftUI-Agent-Skill

    Comprehensively reviews SwiftUI code for best practices on modern APIs, maintainability, and performance.

    5.2k GitHub starsUsed in 2 repos~1.5k tokens
    MobileAuto-check passed
  • Swiftui UI Patterns

    AFK-surf/OpenBridge

    Best practices and example-driven guidance for building SwiftUI views and components.

    430 GitHub starsUsed in 4 repos~887 tokens
    MobileAuto-check passed
  • Swiftui Performance Audit

    harperreed/dotfiles

    Audit and improve SwiftUI runtime performance from code review and architecture.

    334 GitHub starsUsed in 8 repos~1.4k tokens
    MobileAuto-check passed
  • Swift Testing Expert

    AvdLee/Swift-Testing-Agent-Skill

    Expert guidance for Swift Testing: test structure, expect/require macros, traits and tags, parameterized tests, test plans, parallel execution, async waiting patterns, and XCTest migration.

    470 GitHub starsUsed in 1 repo~1.2k tokens
    MobileAuto-check passed
  • Hig Project Context

    raintree-technology/hig-doctor

    Create or update a shared Apple design context document that other HIG skills use to tailor guidance.

    143 GitHub starsUsed in 5 repos~1.2k tokens
    MobileAuto-check passed
  • Swift Testing Pro

    twostraws/Swift-Testing-Agent-Skill

    Writes, reviews, and improves Swift Testing code using modern APIs and best practices.

    453 GitHub stars~1.1k tokensUpdated 4 mo ago
    MobileAuto-check passed

Categories

Questions about Brand Region Client

What does Brand Region Client do?

Add a new brand/region client to BetterBlueKit, or fix errors in an existing one. Brand Region Client is an agent skill from schmidtwmark/BetterBlue. Add a new brand/region client to BetterBlueKit, or fix errors in an existing one.

When should I use Brand Region Client?

Brand Region Client fits situations like: the user asks to add Kia Canada; implement Hyundai Australia; fix the European client.

How do I install Brand Region Client in Claude Code?

Run `npx skills add schmidtwmark/BetterBlue --skill brand-region-client -a claude-code`. Or copy the skill folder (.claude/skills/brand-region-client in schmidtwmark/BetterBlue) into .claude/skills/brand-region-client in your project. Claude Code loads it when a task matches its description.

How do I install Brand Region Client in Codex?

Run `npx skills add schmidtwmark/BetterBlue --skill brand-region-client -a codex`. Or copy the skill folder (.claude/skills/brand-region-client in schmidtwmark/BetterBlue) into .agents/skills/brand-region-client in your project. Codex loads it when a task matches its description.

Can I use Brand Region Client in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add schmidtwmark/BetterBlue --skill brand-region-client -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/brand-region-client, .gemini/skills/brand-region-client, .github/skills/brand-region-client and .opencode/skills/brand-region-client in your project.

What does Brand Region Client need to run?

Going by SKILL.md and its folder, Brand Region Client needs the command-line tools its instructions call (swift).

Does Brand Region Client access the network?

SKILL.md names 1 domain. In commands or code: raw.githubusercontent.com; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Brand Region Client safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Brand Region Client use?

Brand Region Client is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Brand Region Client use?

About 2.1k tokens (SKILL.md is roughly 8.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.3k tokens, read only when the agent opens those files.

What are the alternatives to Brand Region Client?

Skills that share tags, products or a category with Brand Region Client: Swiftui Pro (twostraws/SwiftUI-Agent-Skill, 5.2k stars), Swiftui UI Patterns (AFK-surf/OpenBridge, 430 stars), Swiftui Performance Audit (harperreed/dotfiles, 334 stars) and Swift Testing Expert (AvdLee/Swift-Testing-Agent-Skill, 470 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Brand Region Client?

schmidtwmark (a GitHub user) maintains it in schmidtwmark/BetterBlue, which has 132 GitHub stars. The repository was last updated on October 6, 2026.

Source: schmidtwmark/BetterBlue on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.