Agent skill

Browser4 Coding

by platonai in platonai/Browser4

Create and validate Browser4 plugins, skills, JS scripts, and shell scripts; and develop Browser4 ITSELF (self-development): build modules, extract live skeletons, analyze impact, plan dev tasks…

Apache-2.0Auto-check passedMobile

Install Browser4 Coding

skills CLI
$ npx skills add platonai/Browser4 --skill browser4-coding -a claude-code

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

GitHub CLI
$ gh skill install platonai/Browser4 browser4-coding --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/platonai/Browser4.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/browser4-coding .claude/skills/browser4-coding && 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
browser4-coding
GitHub stars
1.2k
Token cost
~3.4k tokens
SKILL.md length
1,046 words
Files
1
Skills in repo
17
Repo updated
First seen
Licence
Apache-2.0

At a glance

Create and validate Browser4 plugins, skills, JS scripts, and shell scripts; and develop Browser4 ITSELF (self-development): build modules, extract live skeletons, analyze impact, plan dev tasks…

  • Works in 3 steps: Create and validate a new plugin → Copy an existing plugin's structure → Repo governance check before merging
  • The user asks to write a Browser4 plugin
  • SKILL.md covers Quick Start, When to Use, How It Works and Patterns, plus 7 more sections
  • Calls mvn

What it does

Browser4 Coding is an agent skill from platonai/Browser4. Create and validate Browser4 plugins, skills, JS scripts, and shell scripts; and develop Browser4 ITSELF (self-development): build modules, extract live skeletons, analyze impact, plan dev tasks, and check CDP pitfalls. Use when the user asks to write a Browser4 plugin or skill, write browser JS or build scripts, modify Browser4's own Kotlin/Rust code, or validate repo consistency.

Its SKILL.md is about 3.4k 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 Mobile, covering Android development and Shell scripting. It works with Kotlin and Rust. The repository describes itself as: Browser4 — an AI-native browser engine for autonomous agents, intelligent extraction, and large-scale web automation. The licence is Apache-2.0.

When your agent uses it

  • The user asks to write a Browser4 plugin
  • Write browser JS
  • Modify Browser4s own Kotlin/Rust code
  • Validate repo consistency

Example prompts

  • “/browser4-coding”

Requirements

  • Pre-approved tools (allowed-tools): coding.scaffold, coding.scaffoldFlow, coding.scaffoldFromExample, coding.validate, coding.write, coding.read, coding.replace, coding.replaceRegex, coding.editLines, coding.insertAfter, coding.revert, coding.shell, coding.mvnBuild, coding.ktSymbols, coding.ktReferences, coding.ktInheritance, coding.impact, coding.moduleGraph, coding.devTask, coding.trapCheck, coding.protect, coding.tokenStats, coding.estimateTokens, tab.eval, tab.console

Workflow steps

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

  1. Create and validate a new plugin
  2. Copy an existing plugin's structure
  3. Repo governance check before merging

What it can do on your machine

Read from SKILL.md and the folder at commit 0fdba82. 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:

    • coding.scaffold
    • coding.scaffoldFlow
    • coding.scaffoldFromExample
    • coding.validate
    • coding.write
    • coding.read
    • coding.replace
    • coding.replaceRegex
    • coding.editLines
    • coding.insertAfter

    …and 15 more on the same allowed-tools line.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • mvn

    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

Browser4 Coding loads about 3.4k tokens when it runs. Until then it costs about 100 tokens; SKILL.md has 1,046 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~100
When it runs · the whole SKILL.md, loaded when a task matches
~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 platonai/Browser4 at commit 0fdba82, republished under its Apache-2.0 licence (© platonai). 1,046 words, ~3,449 tokens.

Download SKILL.mdSave it as .claude/skills/browser4-coding/SKILL.md (or your agent's skills folder).
name
browser4-coding
description
Create and validate Browser4 plugins, skills, JS scripts, and shell scripts; and develop Browser4 ITSELF (self-development): build modules, extract live skeletons, analyze impact, plan dev tasks, and check CDP pitfalls. Use when the user asks to write a Browser4 plugin or skill, write browser JS or build scripts, modify Browser4's own Kotlin/Rust code, or validate repo consistency.
allowed-tools
coding.scaffold, coding.scaffoldFlow, coding.scaffoldFromExample, coding.validate, coding.write, coding.read, coding.replace, coding.replaceRegex, coding.editLines, coding.insertAfter, coding.revert, coding.shell, coding.mvnBuild, coding.ktSymbols, coding.ktReferences, coding.ktInheritance, coding.impact, coding.moduleGraph, coding.devTask, coding.trapCheck, coding.protect, coding.tokenStats, coding.estimateTokens, tab.eval, tab.console
title
browser4-coding
tier
procedure

browser4-coding

Quick Start

Create and validate a plugin, or a skill:

text
coding.scaffold(type="plugin", pluginName="My Plugin", domain="media",
                basePackage="ai.platon.pulsar.myplugin",
                toolMethod="detectMedia", toolDescription="Detect media on the page")
coding.validate(type="plugin", path="browser4-plugins/browser4-myplugin")

coding.scaffold(type="skill", name="my-skill", description="What it does")
coding.validate(type="skill", path="skills/my-skill/SKILL.md")

For self-development work (modifying Browser4 itself), start with coding.scaffoldFlow — see Workflows below.

The programming-agent kernel for Browser4: sandboxed shell + filesystem, artifact scaffolding/validation (plugins, skills, browser JS, scripts), and a full self-development toolset for modifying Browser4's own code (live module graph, impact analysis, Maven builds with structured diagnostics, CDP pitfall awareness, high-level dev-task planning).

When to Use

  • User wants to create a new Browser4 plugin, skill (SKILL.md), browser JS script, or build/deploy/run script
  • User wants to validate any of the above
  • User wants to modify Browser4's own code (Kotlin modules, Rust CLI, poms) — use the self-development tools below
  • User wants to check repo governance (versions, module registration) or the CDP pitfalls before editing the browser driver

How It Works

The coding tools fall into three layers: artifact tools generate and validate plugins, skills, JS, and scripts (templates from real code so they never go stale); self-development tools inspect and modify Browser4 itself (live module graph, impact analysis, Maven builds, CDP trap checks); and shell/filesystem tools provide a sandboxed environment for anything else.

Patterns

1. Create and validate a new plugin
text
coding.scaffold(type="plugin", pluginName="...", domain="...", basePackage="...", toolMethod="...", toolDescription="...")
coding.validate(type="plugin", path="browser4-plugins/browser4-xxx")
coding.mvnBuild(module="browser4-plugins/browser4-xxx")
2. Copy an existing plugin's structure
text
coding.scaffoldFromExample(path="browser4-plugins/browser4-seo")
coding.validate(type="plugin", path="browser4-plugins/browser4-xxx")
3. Repo governance check before merging
text
coding.validate(type="repo-consistency")

Flags

coding is a tool set, not a CLI — there are no command-line flags. Tool parameters are documented per tool in the sections below.

Errors & Recovery

SymptomCauseFix
coding.validate failsMissing/incorrect required parametersRead the validation output and fix the named field
coding.mvnBuild failsCompile errors in generated codeRead the structured diagnostics and fix the faulty file
scaffoldFromExample renames wrong classesAmbiguous class stemPass explicit per-class keys (e.g. SeoService=CustomService)

Artifact Tools (plugins / skills / js / scripts)

coding.scaffold

Generate a template for any of the four artifact types.

TypeRequired ParametersOptional Parameters
pluginpluginName, domain, basePackage, toolMethod, toolDescriptionpdkVersion (defaults to the repo VERSION file)
skillname (must equal directory name), description (1-1024 chars)triggers (comma-separated), tools (comma-separated → allowed-tools)
jsnamepurpose (extract, inject, interact)
scriptnamescriptType (build, deploy, run), shell (ps1, bash)
coding.validate

Validate an artifact — or the whole repo (type="repo-consistency", no path).

TypePath MeaningWhat It Checks
pluginPlugin base directorypom.xml (parent browser4-pdk, literal version, required deps), plugin.json (PluginManifest schema), AutoConfiguration implements ToolMount, ToolExecutor structure, manifest.name == artifactId
skillSKILL.md file pathYAML frontmatter name + description (1-1024), name == directory name, allowed-tools, tool-reference cross-check
js.js file pathBracket balance, return statement, use strict, anti-patterns
script.ps1/.sh file pathparam block, shebang, error handling
repo-consistency(none)VERSION format; VERSION == root pom == BOM version; every default <module> dir exists; on-disk module dirs registered in the root pom
coding.scaffoldFromExample — anti-staleness live templates

Generate a skeleton from EXISTING real code instead of hand-written templates (which go stale):

  • Single file: parameterizes basePackage/className/domain/toolMethod into {placeholders}; omit rename args to see discovered parameters.
  • Directory (plugin/module): extracts a MULTI-FILE skeleton set with cross-file-consistent renames — basePackage = common package prefix (per-file .tools/.config suffixes preserved), artifactId from pom.xml (parent BOM excluded), pluginName from plugin.json. Renaming className (or passing stem=<NewStem>) derives sibling classes sharing the detected class stem (Seo → Weather renames SeoToolExecutor and SeoAutoConfiguration/SeoService/SeoConfig). Explicit per-class keys (SeoService=CustomService) always win.
coding.scaffoldFlow — multi-file dev-flow skeletons

type: b4-cli-command (commands.rs CommandDef + MCPToolController alias + backend + test), agent-tool (ToolExecutor + ToolMount auto-config), rest-endpoint (Controller + Service + test), test-class, skill. All identifiers derive from one name; verify=true runs the type-appropriate build check.

Self-Development Tools (modifying Browser4 itself)

coding.moduleGraph — live module graph

Scans the repo's real pom.xml files (module path, artifactId, parent, internal deps) and reports the graph; warns when it drifted from the static snapshot (e.g. browser4-pdk, plugins, test modules). module=<path> reports transitive dependents + suggested test commands. This is the anti-staleness answer to a hand-maintained module map.

coding.impact(path=...)

Which module owns a file, which modules are affected transitively (from the LIVE pom graph), and suggested test commands. Run before modifying Browser4's own code.

Show full SKILL.md (415 more words)Show less
coding.mvnBuild(module=..., goals=...)

mvn -pl <module> -am <goals> with structured Kotlin/Java compiler diagnostics instead of raw logs — the fast, dependency-free alternative to a Kotlin LSP server. Use to check code after edits (goals default compile, skipTests=true).

coding.ktSymbols / coding.ktReferences / coding.ktInheritance

Zero-dependency Kotlin analysis: symbols (classes/functions/properties), references (scope='file' default, or scope='module' for a cross-file scan that excludes the declaring file), and the inheritance chain (ktInheritance(path, className?) walks the primary supertype across the module's files). Use before refactoring to assess impact without starting a JDTLS server.

coding.devTask(task=..., verify=..., runTests=...)

High-level entry: parse a natural-language dev task into an executable plan following the AGENTS.md flow — locate files → impact → mvnBuild compile → smallest-scope tests → trapCheck (driver code) → repo-consistency → commit guidance. Module mentions resolve against the LIVE pom graph when available. verify=true runs the fast checks (compile, trapCheck, repo-consistency); runTests=true (with verify) also runs the module's test suite — scoped to -Dtest=<FooTest> when the task names a test class. module= overrides the inferred module.

coding.trapCheck(path=...)

Scan a file for the three known Browser4 CDP pitfalls (AGENTS.md): mouseWheel race (crbug.com/444929150), cursor positioning after focus+click, Input.insertText racing. Run before editing browser-driver code (PulsarWebDriver.kt and friends).

coding.protect(path=..., on=...)

Session-level dynamic file protection: coding.protect(path="src/Foo.kt", on=true) blocks delete/replace/editLines/insertAfter on that exact file; on=false removes it; coding.protect() lists dynamic protections. Repo-governance defaults (VERSION/AGENTS.md/CLAUDE.md/root pom/BOM/CI) are always protected and cannot be unprotected.

Shell & Filesystem Tools

  • coding.shell(command, timeoutSeconds?, workingDir?) — execute allowed dev commands (git/cargo/mvn/npm/python/...)
  • File ops: read, readLines, write, append, replace, replaceRegex, editLines, insertAfter, revert, delete, mkdir, copy, move, stat, diff (myers/patience)
  • Search: glob, grep (skip excluded dirs), languages
  • Repo governance protection: delete/replace/editLines/insertAfter are BLOCKED for VERSION, AGENTS.md, CLAUDE.md, root pom.xml, browser4-dependencies/pom.xml (BOM), .github/workflows/ci.yml — do not try to work around this; ask the user for explicit intent instead. Add session-level protections with coding.protect(path, on).
  • coding.runCode(language, code) — sandboxed snippet execution (kotlin/js/python/bash)
  • LSP: diagnostics(path), symbols(pattern?), references(path, symbol), lspServers — per-language servers (ts/js/py/rs), started on demand, requires the server installed; degrade gracefully when missing
  • coding.tokenStats(reset?) — per-method token usage of coding tool calls so far (calls/errors/input/output tokens, avg/max output); use to audit context-window consumption. coding.estimateTokens(text) — heuristic token count of any text before sending it.

Workflows

Creating a Browser4 Plugin
1. coding.scaffold(type="plugin", pluginName="browser4-xxx", domain="xxx",
     basePackage="ai.platon.pulsar.xxx", toolMethod="doXxx",
     toolDescription="Does xxx on the page")

2. coding.write(path=<file>, content=<from scaffold>) ... write each file

3. coding.validate(type="plugin", path="browser4-plugins/browser4-xxx") → fix issues

4. coding.mvnBuild(module="browser4-plugins/browser4-xxx") → compile to verify

5. If compile fails: coding.read the error file → coding.replace → recompile
Cloning an existing plugin (anti-staleness)
1. coding.scaffoldFromExample(path="browser4-plugins/browser4-seo")
   → Discovery mode: shows basePackage/className/domain/toolMethod/artifactId/stem

2. coding.scaffoldFromExample(path="browser4-plugins/browser4-seo",
     className="WeatherToolExecutor", basePackage="ai.platon.pulsar.weather",
     domain="weather", artifactId="browser4-weather")
   → Multi-file skeleton; sibling classes (WeatherAutoConfiguration/WeatherService)
     follow the stem rename automatically

3. coding.write each generated file → coding.mvnBuild → coding.validate(type="plugin")
1. coding.impact(path=<file>)      → owning module + affected modules + test commands
2. coding.read the file            → understand the code
3. coding.trapCheck(path=<file>)   → CDP pitfalls if browser-driver code
4. Make the edit (coding.replace / coding.editLines / coding.write)
5. coding.mvnBuild(module=<module>) → compile check
6. coding.shell(command=<smallest test scope>) → run tests
7. coding.validate(type="repo-consistency") → governance still consistent
8. coding.moduleGraph() → confirm no module-graph drift
Creating a Skill / JS Script / Build Script
# Skill
1. coding.scaffold(type="skill", name="my-skill", description="...", triggers="...", tools="coding.read,tab.eval")
2. coding.write(path="skills/my-skill/SKILL.md", content=<from scaffold>)  # dir name MUST equal name
3. coding.validate(type="skill", path="skills/my-skill/SKILL.md")

# Browser JS
1. coding.scaffold(type="js", name="extract-prices", purpose="extract")
2. coding.write(path="scripts/extract-prices.js", content=<from scaffold>)
3. coding.validate(type="js", path="scripts/extract-prices.js")
4. tab.eval(expression=<read the JS file>) → tab.console() → iterate

# Build script
1. coding.scaffold(type="script", name="build", scriptType="build", shell="ps1")
2. coding.write(path="build.ps1", content=<from scaffold>)
3. coding.validate(type="script", path="build.ps1") → coding.shell(command="powershell -File build.ps1")

Notes

  • The generated plugin AutoConfiguration implements ToolMount, so PluginManager registers the plugin's tools into CustomToolRegistry automatically.
  • plugin.json follows the real PluginManifest schema (name/version/description/dependsOn/autoConfigurationClasses).
  • The coding module is intentionally independent of pulsar-common; heavy backends (kotlin-compiler-embeddable, LSP servers) are probed at runtime and NEVER downloaded/loaded by default — tools degrade gracefully when unavailable.

© platonai, Apache-2.0. 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 skills/browser4-coding of platonai/Browser4.

Open the folder on GitHubat commit 0fdba82

Compare with similar skills

Browser4 Coding 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.

Browser4 Coding compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Browser4 Coding this skillplatonai/Browser41.2k—~3.4kAutomated safety check: PassApache-2.0
Build Teaql Appteaql/teaql-agent-kit2.8k—~4.6kAutomated safety check: PassMIT
Healthmd CLI QACodyBontecou/health-md230—~4kAutomated safety check: PassAGPL-3.0
Architecture Overclocknekomangaorg/Neko2.8k—~1.4kAutomated safety check: PassApache-2.0
Architecture Renovatornekomangaorg/Neko2.8k—~1.2kAutomated safety check: PassApache-2.0
Publish ReleaseAyuilos/Miffan192—~635Automated safety check: PassAGPL-3.0

Similar skills

  • Build Teaql App

    teaql/teaql-agent-kit

    Build or change a TeaQL application in Java, Rust, Go, Swift, Python, C/.NET, or TypeScript, including Kotlin/JVM applications that consume Java-generated libraries.

    2.8k GitHub stars~4.6k tokensUpdated 10 days ago
    MobileAuto-check passed
  • Healthmd CLI QA

    CodyBontecou/health-md

    Test the standalone Health.md CLI, portable healthmd-mcp server, and direct mobile paths.

    230 GitHub stars~4k tokensUpdated today
    MobileAuto-check passed
  • Architecture Overclock

    nekomangaorg/Neko

    Resolves deep, structural performance bottlenecks in the Kotlin Android codebase.

    2.8k GitHub stars~1.4k tokensUpdated today
    MobileAuto-check passed
  • Architecture Renovator

    nekomangaorg/Neko

    Resolves deep architectural debt in the Kotlin Android codebase through macro-level refactoring.

    2.8k GitHub stars~1.2k tokensUpdated today
    MobileAuto-check passed
  • Publish Release

    Ayuilos/Miffan

    Publish a GitHub release for this fork, with a bilingual changelog that separates fork-owned changes from changes introduced by upstream merges.

    192 GitHub stars~635 tokensUpdated yesterday
    MobileAuto-check passed
  • Code Steward

    nekomangaorg/Neko

    Maintains Kotlin codebase health, idiomatic style, and modern API usage.

    2.8k GitHub stars~977 tokensUpdated today
    MobileAuto-check passed

More from platonai/Browser4

All 17 skills in this repo
  • Data Validation

    platonai/Browser4

    Validates data against common and custom rules (required fields, formats, ranges).

    1.2k GitHub stars~896 tokensUpdated today
    Auto-check passed
  • Form Filling

    platonai/Browser4

    Automatically fills web forms using provided field data and can optionally submit the form.

    1.2k GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • Web Scraping

    platonai/Browser4

    Extracts data from web pages using browser automation and CSS/JavaScript selectors.

    1.2k GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • Organize Task Files

    platonai/Browser4

    Lists, pairs, deduplicates, and moves task files across the Coworker task state machine (0draft → 6git-pushed).

    1.2k GitHub stars~801 tokensUpdated today
    Auto-check passed
  • Task Token Usage

    platonai/Browser4

    Analyzes Claude Code session traces (JSONL) and draft task files to report token usage per task.

    1.2k GitHub stars~701 tokensUpdated today
    Auto-check passed
  • Weather

    platonai/Browser4

    Fetches current weather conditions and a 7-day forecast for a requested location using Open-Meteo geocoding and forecast APIs.

    1.2k GitHub stars~1k tokensUpdated today
    Auto-check passed

Works with

Questions about Browser4 Coding

What does Browser4 Coding do?

Create and validate Browser4 plugins, skills, JS scripts, and shell scripts; and develop Browser4 ITSELF (self-development): build modules, extract live skeletons, analyze impact, plan dev tasks…. Browser4 Coding is an agent skill from platonai/Browser4. Create and validate Browser4 plugins, skills, JS scripts, and shell scripts; and develop Browser4 ITSELF (self-development): build modules, extract live skeletons, analyze impact, plan dev tasks, and check CDP pitfalls.

When should I use Browser4 Coding?

Browser4 Coding fits situations like: the user asks to write a Browser4 plugin; write browser JS; modify Browser4s own Kotlin/Rust code; validate repo consistency.

How do I install Browser4 Coding in Claude Code?

Run `npx skills add platonai/Browser4 --skill browser4-coding -a claude-code`. Or copy the skill folder (skills/browser4-coding in platonai/Browser4) into .claude/skills/browser4-coding in your project. Claude Code loads it when a task matches its description.

How do I install Browser4 Coding in Codex?

Run `npx skills add platonai/Browser4 --skill browser4-coding -a codex`. Or copy the skill folder (skills/browser4-coding in platonai/Browser4) into .agents/skills/browser4-coding in your project. Codex loads it when a task matches its description.

Can I use Browser4 Coding 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 platonai/Browser4 --skill browser4-coding -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/browser4-coding, .gemini/skills/browser4-coding, .github/skills/browser4-coding and .opencode/skills/browser4-coding in your project.

What does Browser4 Coding need to run?

Going by SKILL.md and its folder, Browser4 Coding needs the command-line tools its instructions call (mvn). Its frontmatter pre-approves these tools: coding.scaffold, coding.scaffoldFlow, coding.scaffoldFromExample, coding.validate, coding.write, coding.read, coding.replace, coding.replaceRegex, coding.editLines, coding.insertAfter, coding.revert, coding.shell, coding.mvnBuild, coding.ktSymbols, coding.ktReferences, coding.ktInheritance, coding.impact, coding.moduleGraph, coding.devTask, coding.trapCheck, coding.protect, coding.tokenStats, coding.estimateTokens, tab.eval, tab.console.

Does Browser4 Coding 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 Browser4 Coding 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 Browser4 Coding use?

Browser4 Coding is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Browser4 Coding use?

About 3.4k tokens (SKILL.md is roughly 14k 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 Browser4 Coding?

Skills that share tags, products or a category with Browser4 Coding: Build Teaql App (teaql/teaql-agent-kit, 2.8k stars), Healthmd CLI QA (CodyBontecou/health-md, 230 stars), Architecture Overclock (nekomangaorg/Neko, 2.8k stars) and Architecture Renovator (nekomangaorg/Neko, 2.8k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Browser4 Coding?

platonai (a GitHub user) maintains it in platonai/Browser4, which has 1,152 GitHub stars. The repository holds 17 skills in this directory. The repository was last updated on October 7, 2026.

Source: platonai/Browser4 on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.