Agent skill

Tns Review

by Touch-N-Stars in Touch-N-Stars/Touch-N-Stars

Review changed Touch'N'Stars code with three parallel agents, one each for functionality, UI and maintainability, against this project's own rules (NINA/PINS dual mode, polling contract…

GPL-3.0Auto-check passedFrontend & Design

Install Tns Review

skills CLI
$ npx skills add Touch-N-Stars/Touch-N-Stars --skill tns-review -a claude-code

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

GitHub CLI
$ gh skill install Touch-N-Stars/Touch-N-Stars tns-review --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/Touch-N-Stars/Touch-N-Stars.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/tns-review .claude/skills/tns-review && 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
tns-review
GitHub stars
124
Token cost
~1.7k tokens
SKILL.md length
895 words
Files
1
Skills in repo
7
Repo updated
First seen
Licence
GPL-3.0

At a glance

Review changed Touch'N'Stars code with three parallel agents, one each for functionality, UI and maintainability, against this project's own rules (NINA/PINS dual mode, polling contract…

  • Works in 4 steps: determine the scope → spawn all three agents in one message → merge the findings yourself → …
  • Tasks that involve Subagents
  • SKILL.md covers When this is worth it, Step 1 — determine the scope, Step 2 — spawn all three… and Step 3 — merge the findings…, plus 1 more section
  • Calls git

What it does

Tns Review is an agent skill from Touch-N-Stars/Touch-N-Stars. Review changed Touch'N'Stars code with three parallel agents, one each for functionality, UI and maintainability, against this project's own rules (NINA/PINS dual mode, polling contract, mobile-first UI, i18n). Use before opening a PR or when asked for a thorough review of a branch or working tree.

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Frontend & Design, covering Subagents, Internationalization and Responsive design. The repository describes itself as: a WebApp to controll NINA. The licence is GPL-3.0.

When your agent uses it

  • Tasks that involve Subagents
  • Tasks that involve Internationalization
  • Tasks that involve Responsive design

Example prompts

  • “/tns-review”

Workflow steps

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

  1. determine the scope
  2. spawn all three agents in one message
  3. merge the findings yourself
  4. stop there

What it can do on your machine

Read from SKILL.md and the folder at commit c7af082. 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:

    • git

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

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    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

Tns Review loads about 1.7k tokens when it runs. Until then it costs about 78 tokens; SKILL.md has 895 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~78
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

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 Touch-N-Stars/Touch-N-Stars at commit c7af082, republished under its GPL-3.0 licence (© Touch-N-Stars). 895 words, ~1,741 tokens.

Download SKILL.mdSave it as .claude/skills/tns-review/SKILL.md (or your agent's skills folder).
name
tns-review
description
Review changed Touch'N'Stars code with three parallel agents, one each for functionality, UI and maintainability, against this project's own rules (NINA/PINS dual mode, polling contract, mobile-first UI, i18n). Use before opening a PR or when asked for a thorough review of a branch or working tree.

Three-lens agent review

Runs three read-only review agents in parallel, each with its own lens, then merges their findings. Invoking this skill is the explicit request to spawn subagents.

When this is worth it

Worth it: a feature branch before the PR, a change that touches equipment control, connection handling or a shared store, anything that has to work in both NINA and PINS mode.

Not worth it: a one-line fix, a locale-only change, generated files. Use the built-in /code-review for generic correctness passes — this skill exists for the project-specific rules a generic reviewer cannot know.

Step 1 — determine the scope

bash
git diff --stat                        # working tree
git diff origin/develop...HEAD --stat  # branch, when the working tree is clean

Default: review the working-tree diff. If it is empty, review origin/develop...HEAD. If both are empty, stop and say so — do not invent a scope. An argument (/tns-review <ref|path>) overrides this.

Read the diff yourself before spawning. The agents need a concrete brief, not "review the branch": which files, which layers (component / store / service), which runtime modes are affected, and whether native (Capacitor) code is touched.

Step 2 — spawn all three agents in one message

Use subagent_type: "general-purpose", all three in a single message so they run in parallel. Give each the same scope block, then its own checklist.

Every agent prompt must carry these constraints verbatim:

Read-only review. Do not edit, write or create any file. Do not commit, and do not run builds, installs or anything that mutates the repository — git diff, git show, grep and reading files only. Report findings; do not fix them. Every finding needs a concrete failure scenario (input or state → wrong behavior). A finding you cannot back with one is not a finding — drop it. State for each finding whether you verified it in the code or are inferring it.

Agent 1 — Functionality
  • Both runtime modes. Does this work in NINA and PINS mode? A field that only one stack reports must be feature-detected on the payload, not branched on store.isPINS — the pattern is src/store/cameraStore.js:109 (TempChangeRunning), which falls back to a heuristic when the field is absent.
  • Polling contract. /v2/socket does not deliver every state change, so polling must not be replaced by WebSocket events. Every new or changed poller goes through createPoller (src/utils/poller.js:9) or useBackgroundAwarePolling (src/utils/appLifecycle.js:27) and pauses when the app is backgrounded.
  • Equipment safety. Connection ordering (switch first with its 5 s wait, mount before guider), profile writes via apiService.profileChangeValue, device selection via setProfileDevice (src/utils/equipmentDevices.js) so it survives the next device-list refresh.
  • Races. Stale responses overwriting newer state, missing fetch generations, re-entrancy guards, and operations that take seconds (mount reconnect, driver reload, PHD2) without a loading flag.
  • Error paths. Backend restart, request timeout, a field missing from the payload, an empty device list — what does the user see?
  • i18n keys referenced in the diff exist in src/locales/en.json.
Agent 2 — UI
  • Mobile-first. Narrow viewports, safe areas, orientation changes. Primary controls keep a 48 px touch target (min-h-touch / min-w-touch, defined via --spacing-touch in src/assets/tailwind.css:36).
  • Design system. Use the tns-* utilities (tns-card, tns-btn-primary, tns-btn-secondary, tns-btn-danger, tns-input, tns-select, tns-status-*), not the legacy raw gray/cyan palette still present in older screens.
  • Low-light operation. Contrast and dark-mode readability must not regress; modals and dialogs stay operable.
  • z-index staffing: Modal 40, LoadingOverlay 60, setup wizard 70, modals opened from the wizard 75, LocationSyncModal 76, PINS upgrade overlay and DialogModal 80. Anything teleported to body needs an explicit value.
  • Drag & drop: every <draggable> needs :fallbackOnBody="true"; never forceFallback: true; watch for backdrop-filter on a parent, which creates a containing block for position: fixed and misplaces the ghost on Android Chrome.
  • Text. No hardcoded user-facing strings. Layout survives longer translations — German runs roughly 30% longer than English.
Show full SKILL.md (299 more words)Show less
Agent 3 — Maintainability
  • Reuse before new code. Check src/utils/, src/composables/ and src/services/api/ for something that already does this. Name the existing helper when one exists.
  • Duplication that will drift. The repo's own examples: selectIndi.vue repeats the same driver-change body ten times while setupWizard/IndiDriverSelect.vue does it generically, and getBaseUrl exists twice (api/core.js and apiPinsService.js) with diverging keys. Flag new instances of that shape.
  • Layer separation. Component ↔ store ↔ service. No axios in components, no DOM work in stores.
  • Tests. New logic needs a test, and it belongs on the extracted util rather than on the component — see src/utils/__tests__/equipmentDevices.test.js.
  • Comments in English, explaining why rather than restating the code. Dead code, unused imports, leftover console noise.
  • Generated files (public/whats-new.json, the plugin registry) must not be hand-edited.

Step 3 — merge the findings yourself

The agents report to you, not to the user.

  1. Deduplicate. The same line found through two lenses is one finding with two reasons, not two findings.
  2. Verify what was inferred. Open the file for every finding marked as inferred and drop the ones that do not hold. Agents are confidently wrong often enough that this step is the point of the merge.
  3. Rank by severity: blocker (breaks a session, breaks one runtime mode, loses user data) → should fix → nice to have.
  4. Report as a list of file:line — what breaks, under which conditions, then a verdict: pass / pass with notes / needs changes, matching the reviewer output in docs/AGENT_GUIDELINES.md:230.

Name what was not covered — untested runtime mode, no hardware validation, native platforms not built. A review that hides its blind spots is worse than a short one.

Step 4 — stop there

Report the findings and stop. Do not fix, do not commit. Applying fixes is a separate request, so the user decides which findings are worth acting on.

© Touch-N-Stars, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/tns-review of Touch-N-Stars/Touch-N-Stars.

Open the folder on GitHubat commit c7af082

Compare with similar skills

Tns Review 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.

Tns Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Tns Review this skillTouch-N-Stars/Touch-N-Stars124—~1.7kAutomated safety check: PassGPL-3.0
Dotnet UInovotnyllc/dotnet-artisan232—~1.3kAutomated safety check: PassMIT
Logical Propertiesthedaviddias/Front-End-Checklist74k—~526Automated safety check: PassMIT
Levyra ComposeLUC4N3X/Levyra-deepsound590—~2.2kAutomated safety check: PassGPL-3.0
Maz UIsecondsky/claude-skills227—~4.5kAutomated safety check: PassMIT
UX Create Manifesth0x91b/dev-3.0310—~2.7kAutomated safety check: PassApache-2.0

Similar skills

  • Dotnet UI

    novotnyllc/dotnet-artisan

    Builds .NET UI apps across Blazor (Server, WASM, Hybrid, Auto), MAUI (XAML, MVVM, Shell, Native AOT), Uno Platform (MVUX, Extensions, Toolkit), WPF (.NET 8+, Fluent theme), WinUI 3 (Windows App SDK…

    232 GitHub stars~1.3k tokensUpdated 3 days ago
    Frontend & DesignAuto-check passed
  • Logical Properties

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing stylesheets, component styles, and responsive behavior related to Use CSS logical properties for i18n and RTL support.

    74k GitHub stars~526 tokensUpdated 4 days ago
    Frontend & DesignAuto-check passed
  • Levyra Compose

    LUC4N3X/Levyra-deepsound

    Implement, debug, or review Levyra Jetpack Compose screens, state projections, navigation, animation, lifecycle, accessibility, performance, RTL, localization, edge-to-edge, adaptive layout, and…

    590 GitHub stars~2.2k tokensUpdated yesterday
    Frontend & DesignAuto-check passed
  • Maz UI

    secondsky/claude-skills

    Maz-UI v4 - Modern Vue & Nuxt component library with 50+ standalone components, composables, directives, theming, i18n, and SSR support.

    227 GitHub stars~4.5k tokensUpdated 13 days ago
    Frontend & DesignAuto-check passed
  • UX Create Manifest

    h0x91b/dev-3.0

    Create the initial Product UX Bible for an existing web or full-screen web app by deeply auditing the repository, using sub-agents when available, and generating docs/ux manifests, schemas, budgets…

    310 GitHub stars~2.7k tokensUpdated today
    Frontend & DesignAuto-check passed
  • Collect bounded runtime and debugger evidence for a 1C project through EDT-MCP, including launch, breakpoints, variables, event log, and cleanup.

    296 GitHub stars~967 tokensUpdated 3 days ago
    Frontend & DesignAuto-check passed

More from Touch-N-Stars/Touch-N-Stars

  • API Endpoint

    Touch-N-Stars/Touch-N-Stars

    Add or change an API call between the app and its backends — the Touch'N'Stars plugin server (C), the NINA Advanced API, or the PINS daemon.

    124 GitHub stars~1.3k tokensUpdated yesterday
    Auto-check passed
  • Changelog

    Touch-N-Stars/Touch-N-Stars

    Write or fix a CHANGELOG.md release entry. An agent skill from Touch-N-Stars/Touch-N-Stars.

    124 GitHub stars~725 tokensUpdated yesterday
    Auto-check passed
  • Equipment

    Touch-N-Stars/Touch-N-Stars

    Work on equipment devices — INDI driver selection, device lists, connect/disconnect, and the NINA profile values behind them (camera, mount, focuser, filter wheel, rotator, switch, weather, flat…

    124 GitHub stars~1.5k tokensUpdated yesterday
    Auto-check passed
  • Feature Kickoff

    Touch-N-Stars/Touch-N-Stars

    Turn a rough feature idea into a written goal and testable acceptance criteria before any code is written.

    124 GitHub stars~2k tokensUpdated yesterday
    Auto-check passed
  • I18n

    Touch-N-Stars/Touch-N-Stars

    Add, rename, or remove user-facing strings in src/locales. An agent skill from Touch-N-Stars/Touch-N-Stars.

    124 GitHub stars~1.1k tokensUpdated yesterday
    Auto-check passed
  • Setup Wizard

    Touch-N-Stars/Touch-N-Stars

    Work on the first-run / equipment setup wizard (src/components/setupWizard/) or on location and coordinate handling.

    124 GitHub stars~1.2k tokensUpdated yesterday
    Auto-check passed

Questions about Tns Review

What does Tns Review do?

Review changed Touch'N'Stars code with three parallel agents, one each for functionality, UI and maintainability, against this project's own rules (NINA/PINS dual mode, polling contract…. Tns Review is an agent skill from Touch-N-Stars/Touch-N-Stars. Review changed Touch'N'Stars code with three parallel agents, one each for functionality, UI and maintainability, against this project's own rules (NINA/PINS dual mode, polling contract, mobile-first UI, i18n).

When should I use Tns Review?

Tns Review fits situations like: tasks that involve Subagents; tasks that involve Internationalization; tasks that involve Responsive design.

How do I install Tns Review in Claude Code?

Run `npx skills add Touch-N-Stars/Touch-N-Stars --skill tns-review -a claude-code`. Or copy the skill folder (.claude/skills/tns-review in Touch-N-Stars/Touch-N-Stars) into .claude/skills/tns-review in your project. Claude Code loads it when a task matches its description.

How do I install Tns Review in Codex?

Run `npx skills add Touch-N-Stars/Touch-N-Stars --skill tns-review -a codex`. Or copy the skill folder (.claude/skills/tns-review in Touch-N-Stars/Touch-N-Stars) into .agents/skills/tns-review in your project. Codex loads it when a task matches its description.

Can I use Tns Review 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 Touch-N-Stars/Touch-N-Stars --skill tns-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/tns-review, .gemini/skills/tns-review, .github/skills/tns-review and .opencode/skills/tns-review in your project.

What does Tns Review need to run?

Going by SKILL.md and its folder, Tns Review needs the command-line tools its instructions call (git).

Does Tns Review access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Tns Review 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 Tns Review use?

Tns Review is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Tns Review use?

About 1.7k tokens (SKILL.md is roughly 7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Tns Review?

Skills that share tags, products or a category with Tns Review: Dotnet UI (novotnyllc/dotnet-artisan, 232 stars), Logical Properties (thedaviddias/Front-End-Checklist, 74k stars), Levyra Compose (LUC4N3X/Levyra-deepsound, 590 stars) and Maz UI (secondsky/claude-skills, 227 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Tns Review?

Touch-N-Stars (a GitHub organization) maintains it in Touch-N-Stars/Touch-N-Stars, which has 124 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on October 10, 2026.

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