Agent skill

Engineering

by wellwelwel in wellwelwel/poku

Engineering deep-dive for poku covering performance, code, and security patterns, TypeScript config, build pipeline, developer experience, and CI/CD.

MITAuto-check passedDevOps & Cloud

Install Engineering

skills CLI
$ npx skills add wellwelwel/poku --skill engineering -a claude-code

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

GitHub CLI
$ gh skill install wellwelwel/poku engineering --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/wellwelwel/poku.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/engineering .claude/skills/engineering && 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
engineering
GitHub stars
1.2k
Token cost
~1.4k tokens
SKILL.md length
710 words
Files
1
Skills in repo
6
Repo updated
First seen
Licence
MIT

At a glance

Engineering deep-dive for poku covering performance, code, and security patterns, TypeScript config, build pipeline, developer experience, and CI/CD.

  • Works in 7 steps: Performance Patterns → Code Patterns → Security Patterns → …
  • Optimizing code
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Writing performance-sensitive logic

What it does

Engineering is an agent skill from wellwelwel/poku. Engineering deep-dive for poku covering performance, code, and security patterns, TypeScript config, build pipeline, developer experience, and CI/CD. Use when implementing or optimizing code, writing performance-sensitive logic, touching the build, or applying code-style and security rules.

Its SKILL.md is about 1.4k 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 DevOps & Cloud, covering CI/CD. It works with TypeScript. The repository describes itself as: 🐷 Poku makes testing easy for Node.js, Bun, Deno, and you at the same time. The licence is MIT.

When your agent uses it

  • Optimizing code
  • Writing performance-sensitive logic
  • Touching the build
  • Applying code-style and security rules

Example prompts

  • “/engineering”

Workflow steps

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

  1. Performance Patterns
  2. Code Patterns
  3. Security Patterns
  4. TypeScript Configuration
  5. Build Pipeline
  6. Developer Experience (DX)
  7. CI/CD

What it can do on your machine

Read from SKILL.md and the folder at commit e02df0b. 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

    No scripts in the folder and no shell commands in SKILL.md.

    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.

Context cost

Engineering loads about 1.4k tokens when it runs. Until then it costs about 76 tokens; SKILL.md has 710 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~76
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

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 wellwelwel/poku at commit e02df0b, republished under its MIT licence (© wellwelwel). 710 words, ~1,399 tokens.

Download SKILL.mdSave it as .claude/skills/engineering/SKILL.md (or your agent's skills folder).
name
engineering
description
Engineering deep-dive for poku covering performance, code, and security patterns, TypeScript config, build pipeline, developer experience, and CI/CD. Use when implementing or optimizing code, writing performance-sensitive logic, touching the build, or applying code-style and security rules.

Engineering patterns, performance, security, and build/CI. Read the actual source and config to confirm details, since implementations change over time.

1. Performance Patterns

Look in src/services/ and src/polyfills/:

  • Each test file runs in its own process via spawn with shell: false. stdout/stderr are piped. A plugin can request an extra IPC channel via plugin.ipc
  • The runner uses a concurrency-bounded promise queue. The default limit is availableParallelism(). Sequential mode or isolation: none forces a limit of 1
  • availableParallelism is polyfilled by feature-detecting the native function at runtime and falling back to the CPU count

Authoring rules to respect when writing performance-sensitive code:

  • NEVER let extra or uncommon features compromise or pollute the default path. Keep them behind a user-triggered conditional so the code stays unreached ("dead code") until the flag activates it
  • ALWAYS early-return to the extreme. The moment a path is decided, leave it with return, break, or continue before doing any further work, so the common path runs the least code
  • Declare each variable ONLY after the guards that could exit, in the narrowest scope that uses it. NEVER allocate on a path that never reaches it
  • ALWAYS cache repeated state checks in the GLOBAL singleton instead of recomputing them

Performance priority order when trade-offs arise:

  1. Long-term maintainability and readable code
  2. V8 and JIT optimizations, including minimal memory allocation
  3. Algorithmic complexity (Big O)

Performance and maintainability must stay in balance with each other.

2. Code Patterns
  • Functional composition: Object.assign(coreFn, { todo, skip, only }) exposes modifiers on it and describe
  • Fluent formatter: the Formatter class implements [Symbol.toPrimitive]() for implicit string coercion
  • Promise coordination: a captured resolve callback, no external queue dependency
  • GLOBAL singleton: a single object holds runtime config and shared state
  • Parent→Child: POKU_* env vars carry context to child processes
  • Timing: hrtime() (the [seconds, nanoseconds] tuple form) for high-precision durations
  • Naming: NEVER use abbreviated variable names, including basic loop counters. Use index instead of i, element instead of el. Names MUST be fully spelled out and descriptive
  • Comments: AVOID them. Clear code, organization, logic, and naming MUST make them unnecessary. The only exception is a public API exposed directly to the end user, which MAY carry objective inline docs
3. Security Patterns
  • shell: false in spawn() prevents shell injection
  • escapeRegExp() sanitizes user-supplied filter and pattern input before it becomes a RegExp
  • Config auto-discovery only loads an allowlisted set of filenames (an explicit --config= path bypasses the list)
  • Windows path safety: the config loader loads from a file:// URL on Windows, and the in-process test loader uses pathToFileURL()
  • Process kill uses kill -9 on Unix and taskkill /F /T /PID on Windows
Show full SKILL.md (276 more words)Show less
4. TypeScript Configuration

tsconfig.json is the source of truth. The constraint that drives it: output must run on the oldest supported Node, so prefer compatibility over newer language/output targets when changing it.

5. Build Pipeline

scripts/build.sh (wired to the build script in package.json) is the source of truth. It compiles source and tests, then runs post-build fixups. Read it before changing build behavior rather than relying on a description here.

6. Developer Experience (DX)

ALWAYS preserve or improve the developer experience. Read the assertion builder in src/builders/, the formatter and reporters in src/services/, and the runner/output/options parsers in src/parsers/. Key principles:

  • Assertion failures must be actionable: file path, code line, operator, and full actual/expected diff (red/green). Non-comparison assertions (throws, rejects) pass hideDiff: true to keep the output focused
  • Visual hierarchy: status icons, colors, and dynamic indentation that tracks describe/it nesting depth encode test state. Read the reporter and the formatter for the current icon and color mapping
  • Output filtering: in non-debug mode only ANSI-colored lines pass through, hiding noisy library output. Debug mode (or a failed file) passes everything through
  • Zero friction: TypeScript is auto-detected by file extension (the runner injects --import=tsx for .ts/.mts/.cts), config files are auto-discovered, and defineConfig() enables IDE autocomplete
  • Config auto-discovery order: poku.config.js, then .pokurc.json, then .pokurc.jsonc, then built-in defaults
  • Formatter usage: ALWAYS build output with the fluent format('text').bold().success() chain, NEVER raw ANSI codes
7. CI/CD

Read .github/workflows/ for the current pipeline. What constrains the work:

  • Lint must pass (Biome + Prettier, see biome.jsonc)
  • Coverage has a minimum threshold (checkCoverage in .nycrc, the source of truth)
  • Cross-runtime and multi-version compatibility is enforced. See the workflows and test/__docker__/node/ for the current runtimes and versions

© wellwelwel, MIT. 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/engineering of wellwelwel/poku.

Open the folder on GitHubat commit e02df0b

Compare with similar skills

Engineering 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.

Engineering compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Engineering this skillwellwelwel/poku1.2k—~1.4kAutomated safety check: PassMIT
Avotokyo Workflowsheyq02/simpleicons.dev171—~1.5kAutomated safety check: PassApache-2.0
Configuration GeneratorArabelaTso/Skills-4-SE253—~2.8kAutomated safety check: NotesApache-2.0
Workflow Setupathola/claude-night-market342—~1.4kAutomated safety check: PassMIT
Datadog Appdatadog-labs/agent-skills177—~744Automated safety check: PassMIT
Nx Monorepogiuseppe-trisciuoglio/developer-kit355—~2.3kAutomated safety check: NotesMIT

Similar skills

  • Avotokyo Workflows

    heyq02/simpleicons.dev

    Use avotokyo/workflows reusable GitHub Actions for TypeScript/Vite+ projects.

    171 GitHub stars~1.5k tokensUpdated 18 days ago
    DevOps & CloudAuto-check passed
  • Configuration Generator

    ArabelaTso/Skills-4-SE

    Generate configuration files for applications, services, and infrastructure.

    253 GitHub stars~2.8k tokensUpdated 1 mo ago
    DevOps & CloudAuto-check: notes
  • Workflow Setup

    athola/claude-night-market

    Configures GitHub Actions CI/CD workflows for testing, linting, and deployment.

    342 GitHub stars~1.4k tokensUpdated 2 days ago
    DevOps & CloudAuto-check passed
  • Datadog App

    datadog-labs/agent-skills

    Guides developers building Datadog Apps with TypeScript, React, the @datadog/apps scaffolder, and @datadog/vite-plugin.

    177 GitHub stars~744 tokensUpdated yesterday
    DevOps & CloudAuto-check passed
  • Nx Monorepo

    giuseppe-trisciuoglio/developer-kit

    Provides comprehensive Nx monorepo management guidance for TypeScript/JavaScript projects.

    355 GitHub stars~2.3k tokensUpdated 28 days ago
    DevOps & CloudAuto-check: notes
  • Linea Dependency Maintenance

    Consensys-Incorporated/linea-attestation-registry

    Safely plan and execute dependency maintenance for JavaScript/TypeScript (npm, pnpm) and GitHub Actions, including npm lockfiles, pnpm workspaces, catalogs, overrides, SHA-pinned action versions…

    177 GitHub starsUsed in 1 repo~3.7k tokens
    DevelopmentAuto-check: warnings

More from wellwelwel/poku

  • Benchmarking

    wellwelwel/poku

    Empirical performance comparison playbook for poku covering hypothesis lenses, a standalone harness with a byte-identical correctness gate, variant generation, sequential hyperfine measurement…

    1.2k GitHub stars~3k tokensUpdated 3 mo ago
    Auto-check passed
  • Architecture

    wellwelwel/poku

    Architecture deep-dive for poku covering project structure, execution flow, config discovery, runtime detection, and competitive context.

    1.2k GitHub stars~867 tokensUpdated 3 mo ago
    Auto-check passed
  • Documentation

    wellwelwel/poku

    Documentation deep-dive for the poku website covering Docusaurus setup, language policy, feature-to-docs mapping, conventions, JSDoc, and the changelog.

    1.2k GitHub stars~982 tokensUpdated 3 mo ago
    Auto-check passed
  • Pre Commit

    wellwelwel/poku

    Run the mandatory pre-commit checks for the poku repository before staging a commit.

    1.2k GitHub stars~728 tokensUpdated 3 mo ago
    Auto-check passed
  • Testing

    wellwelwel/poku

    Testing deep-dive for poku covering test structure, commands, patterns, fixtures, utils, Docker compatibility, and coverage.

    1.2k GitHub stars~1k tokensUpdated 3 mo ago
    Auto-check passed

Works with

Categories

Questions about Engineering

What does Engineering do?

Engineering deep-dive for poku covering performance, code, and security patterns, TypeScript config, build pipeline, developer experience, and CI/CD. Engineering is an agent skill from wellwelwel/poku. Engineering deep-dive for poku covering performance, code, and security patterns, TypeScript config, build pipeline, developer experience, and CI/CD.

When should I use Engineering?

Engineering fits situations like: optimizing code; writing performance-sensitive logic; touching the build; applying code-style and security rules.

How do I install Engineering in Claude Code?

Run `npx skills add wellwelwel/poku --skill engineering -a claude-code`. Or copy the skill folder (.claude/skills/engineering in wellwelwel/poku) into .claude/skills/engineering in your project. Claude Code loads it when a task matches its description.

How do I install Engineering in Codex?

Run `npx skills add wellwelwel/poku --skill engineering -a codex`. Or copy the skill folder (.claude/skills/engineering in wellwelwel/poku) into .agents/skills/engineering in your project. Codex loads it when a task matches its description.

Can I use Engineering 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 wellwelwel/poku --skill engineering -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/engineering, .gemini/skills/engineering, .github/skills/engineering and .opencode/skills/engineering in your project.

What does Engineering need to run?

SKILL.md names no scripts, command-line tools or credentials: Engineering is instructions for the agent only.

Does Engineering 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 Engineering 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 Engineering use?

Engineering 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 Engineering use?

About 1.4k tokens (SKILL.md is roughly 5.6k 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 Engineering?

Skills that share tags, products or a category with Engineering: Avotokyo Workflows (heyq02/simpleicons.dev, 171 stars), Configuration Generator (ArabelaTso/Skills-4-SE, 253 stars), Workflow Setup (athola/claude-night-market, 342 stars) and Datadog App (datadog-labs/agent-skills, 177 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Engineering?

wellwelwel (a GitHub user) maintains it in wellwelwel/poku, which has 1,183 GitHub stars. The repository holds 6 skills in this directory. The repository was last updated on June 18, 2026.

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