New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Unity Development Loop

by akiojin in akiojin/unity-cli

Run Unity runtime development loops with gameplay-focused implementation and validation.

MITAuto-check passedGame Development

Install Unity Development Loop

skills CLI
$ npx skills add akiojin/unity-cli --skill unity-development-loop -a claude-code

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

GitHub CLI
$ gh skill install akiojin/unity-cli unity-development-loop --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/akiojin/unity-cli.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude-plugin/plugins/unity-cli/skills/unity-development-loop .claude/skills/unity-development-loop && 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
unity-development-loop
GitHub stars
107
Token cost
~1.6k tokens
SKILL.md length
780 words
Files
3 (incl. references)
Skills in repo
28
Repo updated
First seen
Licence
MIT

At a glance

Run Unity runtime development loops with gameplay-focused implementation and validation.

  • Works in 8 steps: Confirm the bridge answers unity-cli… → Lock the scenario as observable runtime… → Choose the smallest next Unity-side… → …
  • The user asks to iterate on runtime behavior
  • SKILL.md covers Use When, Do Not Use When, Preferred Flow and Latency regression checks, plus 2 more sections
  • Calls python3

What it does

Unity Development Loop is an agent skill from akiojin/unity-cli. Run Unity runtime development loops with gameplay-focused implementation and validation. Use when the user asks to iterate on runtime behavior, reproduce and fix a gameplay bug, measure Editor operation latency, or implement a Unity-side flow until acceptance criteria are met. Do not use for .inputactions authoring, read-only inspection, or Rust CLI-only work.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/development-loop-playbook.md` and `references/runtime-checklist.md`). Compatibility notes: Requires unity-cli connected to a Unity Editor that supports Play Mode control, runtime input simulation, UI automation, and editor diagnostics.

It sits in Game Development, covering User stories. It works with Rust. The repository describes itself as: Rust CLI for Unity Editor automation over Unity TCP protocol. The licence is MIT.

When your agent uses it

  • The user asks to iterate on runtime behavior
  • Reproduce and fix a gameplay bug
  • Measure Editor operation latency
  • Implement a Unity-side flow until acceptance criteria are met

Example prompts

  • “/unity-development-loop”

Requirements

  • Python 3
  • Compatibility (from SKILL.md): Requires unity-cli connected to a Unity Editor that supports Play Mode control, runtime input simulation, UI automation, and editor diagnostics.
  • Pre-approved tools (allowed-tools): Bash(unity-cli:*), Read, Grep, Glob

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Confirm the bridge answers unity-cli system ping. If it does not, run unity-cli doctor --output json and recover from its diagnosis first…
  2. Lock the scenario as observable runtime behavior and define 1-3 acceptance checks before editing or running anything.
  3. Choose the smallest next Unity-side change and delegate implementation to unity-csharp-edit when code must change.
  4. When only the body of an existing method changes and the Play session state is worth keeping, preview it without leaving Play Mode: check…
  5. Run the narrowest runtime check through unity-playmode-testing, unity-ui-automation, or unity-editor-tools, depending on whether the loop…
  6. Capture only the evidence needed for the current hypothesis: state reads or logs for non-visual behavior, a screenshot for visible end…
  7. Record the iteration with scenario, acceptance criteria, change made, execution, evidence, observations, result, and next action.
  8. Stop when the current evidence satisfies every acceptance check; otherwise continue with one concrete next hypothesis.

What it can do on your machine

Read from SKILL.md and the folder at commit 7f45d7a. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Bash(unity-cli:*)
    • Read
    • Grep
    • Glob

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • python3

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

  • Network

    No URLs in SKILL.md.

    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.

  • Compatibility

    Requires unity-cli connected to a Unity Editor that supports Play Mode control, runtime input simulation, UI automation, and editor diagnostics.

    From compatibility in the SKILL.md frontmatter.

Context cost

Unity Development Loop loads about 1.6k tokens when it runs, and up to ~3.6k if it reads all its reference files. Until then it costs about 97 tokens; SKILL.md has 780 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~97
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3.6k

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 akiojin/unity-cli at commit 7f45d7a, republished under its MIT licence (© akiojin). 780 words, ~1,639 tokens.

Download SKILL.mdSave it as .claude/skills/unity-development-loop/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
unity-development-loop
description
Run Unity runtime development loops with gameplay-focused implementation and validation. Use when the user asks to iterate on runtime behavior, reproduce and fix a gameplay bug, measure Editor operation latency, or implement a Unity-side flow until acceptance criteria are met. Do not use for `.inputactions` authoring, read-only inspection, or Rust CLI-only work.
allowed-tools
Bash(unity-cli:*), Read, Grep, Glob
compatibility
Requires unity-cli connected to a Unity Editor that supports Play Mode control, runtime input simulation, UI automation, and editor diagnostics.
metadata.author
akiojin
metadata.version
0.3.0
metadata.category
testing
metadata.triggers
runtime, gameplay, iterate, acceptance, latency, benchmark

Unity Development Loop

With --output json, results use {success, command, data, errors, warnings}. Check the exit status and envelope success first; tool-result fields in this skill are relative to data. Read failure codes from errors[0].code; see unity-cli-usage for exit-code recovery.

Run Unity-side development as short, acceptance-driven loops. This skill chooses the smallest next change, the narrowest runtime check, and the lightest evidence set that can prove or disprove the current hypothesis.

Use When

  • The user wants to implement and verify a Unity gameplay, input, or UI scenario.
  • The task needs a repeatable loop of Unity-side code changes, Play Mode execution, evidence capture, and runtime confirmation.
  • The user asks to iterate on a runtime bug until acceptance criteria are satisfied.
  • The request needs screenshots, short video, console inspection, or profiler data as part of Unity runtime validation.
  • The user wants to measure Editor operation latency or check performance budgets in a unity-cli source checkout.

Do Not Use When

  • The task is authoring .inputactions assets instead of validating runtime behavior. Use unity-input-system.
  • The task is read-only code or scene investigation with no planned change. Use the relevant inspection skill.
  • The task is Rust CLI-only or does not require Unity runtime validation.
  • The task is scene, prefab, or asset authoring without a runtime validation loop.

Preferred Flow

  1. Confirm the bridge answers unity-cli system ping. If it does not, run unity-cli doctor --output json and recover from its diagnosis first; on SAFE_MODE, fix the reported compile errors (file, line) before any runtime check.
  2. Lock the scenario as observable runtime behavior and define 1-3 acceptance checks before editing or running anything.
  3. Choose the smallest next Unity-side change and delegate implementation to unity-csharp-edit when code must change.
  4. When only the body of an existing method changes and the Play session state is worth keeping, preview it without leaving Play Mode: check unity-cli raw hot_reload_status --json '{}' and, if supported is true, follow the Hot Reload Preview Loop in the playbook. Otherwise edit the file and re-enter Play Mode.
  5. Run the narrowest runtime check through unity-playmode-testing, unity-ui-automation, or unity-editor-tools, depending on whether the loop is gameplay, UI, or diagnostics driven.
  6. Capture only the evidence needed for the current hypothesis: state reads or logs for non-visual behavior, a screenshot for visible end state, short video for timing, and profiler data only for performance questions.
  7. Record the iteration with scenario, acceptance criteria, change made, execution, evidence, observations, result, and next action.
  8. Stop when the current evidence satisfies every acceptance check; otherwise continue with one concrete next hypothesis.
Show full SKILL.md (355 more words)Show less

Latency regression checks

For performance work in a unity-cli source checkout, build the release CLI and run python3 scripts/e2e-matrix.py --suites perf --unity-cli target/release/unity-cli. Select Editors with repeated --editor arguments; release acceptance requires 6000.3.25f1 and 2022.3.62f3 with the default --perf-focus both. This launches isolated GUI projects and runs scripts/bench-editor-ops.py for the staff report's 23 operations, 30 samples after 3 warmup cycles per focus condition. It also runs the existing frontmost editor_eval budget with 100 samples and history. Use an idle host and coordinate Editor focus with other agents. The owned Game Views temporarily use PlayUnfocused; the benchmark restores their previous window settings after measurement so Play does not invalidate background samples. Background screenshots start with Finder frontmost but may activate the target Editor through the existing GameView.Focus() call. JSON records this exception; a third-party PID change or any other operation's focus change invalidates the cycle.

Inspect matrix.json, each perf.json and perf.log. Report the command, Unity version, pass/fail counts and violations. perf-budgets.json gates p50/p95; .unity/perf/editor-ops-history.jsonl also gates p50 degradation greater than 20% against the last five complete matching runs. UNITY_CLI_PERF_REGRESSION_PERCENT overrides that relative threshold for investigation; releases use the default. Read Benchmark Policy in docs/development.md for baseline recovery and conditions.

Timing includes each CLI process startup while Editor/unityd remain warm. The 22 remote operations use unityd. C# file reads use the local reader; Play/Stop include state confirmation. Do not equate these measurements with persistent-shell timings or compare foreground and background as if they were the same condition. After a product/benchmark change, re-run the matrix; a historical PASS is insufficient. For Unity-free transport regressions use python3 scripts/bench-cli-latency.py (CI's required latency gate).

Examples

  • "Implement a jump input tweak, run it in Play Mode, and keep iterating until the jump timing feels correct."
  • "Fix this settings panel flow, click through the UI, capture proof, and stop when the final state matches the spec."
  • "Reproduce the runtime bug, inspect logs, make the smallest fix, and rerun only the affected path."

References

  • runtime-checklist.md: connection, instance, unity-cli doctor recovery, and evidence-capture baseline before starting a loop.
  • development-loop-playbook.md: scenario-specific loop recipes (including the Play Mode hot reload preview), evidence selection rules, and exit criteria.

© akiojin, 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-plugin/plugins/unity-cli/skills/unity-development-loop of akiojin/unity-cli.

  • SKILL.md
  • references/development-loop-playbook.md
  • references/runtime-checklist.md

Open the folder on GitHubat commit 7f45d7a

Compare with similar skills

Unity Development Loop 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.

Unity Development Loop compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Unity Development Loop this skillakiojin/unity-cli107—~1.6kAutomated safety check: PassMIT
Ralph Tui Create Beads Rustsubsy/ralph-tui2.5k1 repos~2.8kAutomated safety check: PassMIT
Create A Plandojoengine/dojo478—~1.7kAutomated safety check: PassApache-2.0
ReleaseBloom-Engine/engine134—~1.4kAutomated safety check: NotesMIT
Rocketmq Rust Good First Issuemxsm/rocketmq-rust1.5k—~2.3kAutomated safety check: PassApache-2.0
Add ThemeIvanWng97/pixtuoid489—~449Automated safety check: PassMIT

Similar skills

  • Ralph Tui Create Beads Rust

    subsy/ralph-tui

    Convert PRDs to beads for ralph-tui execution using beads-rust (br CLI).

    2.5k GitHub starsUsed in 1 repo~2.8k tokens
    Product & Project ManagementAuto-check passed

  • Create A Plan

    dojoengine/dojo

    Conduct a focused technical planning interview to produce an implementable, parallelizable plan or spec with clear dependencies, risks, and open questions.

    478 GitHub stars~1.7k tokensUpdated 5 mo ago
    Game DevelopmentAuto-check passed

  • Release

    Bloom-Engine/engine

    Cut a new Bloom Engine release — bump package.json if requested, commit, tag, push, create GitHub Release, and watch the gated release pipeline

    134 GitHub stars~1.4k tokensUpdated 26 days ago
    Game DevelopmentAuto-check: notes

  • Rocketmq Rust Good First Issue

    mxsm/rocketmq-rust

    Help the rocketmq-rust repository owner or maintainer prepare and publish good first issues for new contributors to claim, with exact files, concrete changes, acceptance criteria, labels, and…

    1.5k GitHub stars~2.3k tokensUpdated today
    Product & Project ManagementAuto-check passed

  • Add Theme

    IvanWng97/pixtuoid

    Add a new color theme to pixtuoid (a full Theme palette, rendered into the office).

    489 GitHub stars~449 tokensUpdated today
    Game DevelopmentAuto-check passed

  • Build Acceleration

    mohitmishra786/low-level-dev-skills

    Build acceleration skill for C/C++ projects. An agent skill from mohitmishra786/low-level-dev-skills.

    253 GitHub stars~1.4k tokensUpdated 3 mo ago
    Game DevelopmentAuto-check passed

More from akiojin/unity-cli

All 28 skills in this repo

  • Gh PR

    akiojin/unity-cli

    Create or update GitHub Pull Requests with the gh CLI, including deciding whether to create a new PR or only push based on existing PR merge status.

    107 GitHub stars~2.7k tokensUpdated 2 days ago
    Auto-check passed

  • Gh Skills Sync

    akiojin/unity-cli

    Sync the gh- automation skills from the upstream skills repository into Codex and project skill directories.

    107 GitHub stars~680 tokensUpdated 2 days ago
    Auto-check: notes

  • Unity 2D Sprite Tilemap

    akiojin/unity-cli

    Build and verify a 2D sprite and Tilemap scene with unity-cli.

    107 GitHub stars~1.6k tokensUpdated 2 days ago
    Auto-check passed

  • Unity Addressables

    akiojin/unity-cli

    Manage Unity Addressables groups and content with unity-cli.

    107 GitHub stars~700 tokensUpdated 2 days ago
    Auto-check passed

  • Unity Asset Management

    akiojin/unity-cli

    Manage Unity assets and import metadata with unity-cli. An agent skill from akiojin/unity-cli.

    107 GitHub stars~1.8k tokensUpdated 2 days ago
    Auto-check passed

  • Unity Audio Setup

    akiojin/unity-cli

    Configure Unity audio with unity-cli. An agent skill from akiojin/unity-cli.

    107 GitHub stars~1.4k tokensUpdated 2 days ago
    Auto-check passed

Works with

Categories

Questions about Unity Development Loop

What does Unity Development Loop do?

Run Unity runtime development loops with gameplay-focused implementation and validation. Unity Development Loop is an agent skill from akiojin/unity-cli. Run Unity runtime development loops with gameplay-focused implementation and validation.

When should I use Unity Development Loop?

Unity Development Loop fits situations like: the user asks to iterate on runtime behavior; reproduce and fix a gameplay bug; measure Editor operation latency; implement a Unity-side flow until acceptance criteria are met.

How do I install Unity Development Loop in Claude Code?

Run `npx skills add akiojin/unity-cli --skill unity-development-loop -a claude-code`. Or copy the skill folder (.claude-plugin/plugins/unity-cli/skills/unity-development-loop in akiojin/unity-cli) into .claude/skills/unity-development-loop in your project. Claude Code loads it when a task matches its description.

How do I install Unity Development Loop in Codex?

Run `npx skills add akiojin/unity-cli --skill unity-development-loop -a codex`. Or copy the skill folder (.claude-plugin/plugins/unity-cli/skills/unity-development-loop in akiojin/unity-cli) into .agents/skills/unity-development-loop in your project. Codex loads it when a task matches its description.

Can I use Unity Development Loop 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 akiojin/unity-cli --skill unity-development-loop -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/unity-development-loop, .gemini/skills/unity-development-loop, .github/skills/unity-development-loop and .opencode/skills/unity-development-loop in your project.

What does Unity Development Loop need to run?

Going by SKILL.md and its folder, Unity Development Loop needs the command-line tools its instructions call (python3). Our summary lists: Python 3. Its frontmatter pre-approves these tools: Bash(unity-cli:*), Read, Grep, Glob. Compatibility (from SKILL.md): Requires unity-cli connected to a Unity Editor that supports Play Mode control, runtime input simulation, UI automation, and editor diagnostics..

Does Unity Development Loop access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Unity Development Loop 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 Unity Development Loop use?

Unity Development Loop 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 Unity Development Loop use?

About 1.6k tokens (SKILL.md is roughly 6.6k 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 1.9k tokens, read only when the agent opens those files.

What are the alternatives to Unity Development Loop?

Skills that share tags, products or a category with Unity Development Loop: Ralph Tui Create Beads Rust (subsy/ralph-tui, 2.5k stars), Create A Plan (dojoengine/dojo, 478 stars), Release (Bloom-Engine/engine, 134 stars) and Rocketmq Rust Good First Issue (mxsm/rocketmq-rust, 1.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Unity Development Loop?

akiojin (a GitHub user) maintains it in akiojin/unity-cli, which has 107 GitHub stars. The repository holds 28 skills in this directory. The repository was last updated on October 5, 2026.

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