apex_skill

This skill orchestrates Analyze → Plan → Execute → eLicit → eXamine to streamline features, fixes, and refactors with automated safety checks.
  • TypeScript

2

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill fusengine/agents --skill apex

  • SKILL.md12.7 KB

Overview

This skill defines the APEX Methodology: a disciplined development workflow (Analyze → Plan → Execute → eLicit → eXamine) that applies to features, bug fixes, refactors and hotfixes. It auto-detects common project types (Laravel, Next.js, React, Swift) and loads framework-specific references while enforcing strict quality rules like file size limits, interface separation, SOLID principles, and expert self-review. Initialization of APEX tracking is mandatory before any write/edit actions.

How this skill works

On trigger words (implement, create, build, fix, add feature, refactor, develop) the skill initializes APEX tracking and runs the workflow phases: parallel analysis agents, task planning with TaskCreate, guided execution with file/interface rules, an expert self-review (eLicit), and a sniper validation pass (eXamine). It auto-selects framework references by inspecting project files (composer.json, next.config.*, vite.config.*, Package.swift) and blocks write/edit until documentation consultation is recorded.

When to use it

  • Starting any new feature, refactor, bug fix, or urgent hotfix
  • When you need a repeatable, auditable dev workflow with enforced code quality
  • For repo onboarding to ensure consistent branching, commits, and PRs
  • When you require framework-specific guidance and toolchain checks
  • Before creating a pull request to ensure sniper validation and tests pass

Best practices

  • Run the APEX tracking initializer before any other action
  • Always run two analysis agents in parallel: explore-codebase and research-expert
  • Keep files <100 lines; split at 90 lines and target 50–80 lines
  • Create interfaces before implementation and place them in src/interfaces or src/types
  • Use TaskCreate to break work into atomic tasks and monitor file sizes
  • Run sniper validation (ZERO errors) and full test suite before PR

Example use cases

  • Implement a new login feature in Next.js with TaskCreate plan, interfaces, unit tests and sniper validation
  • Fix a production bug: write failing test first, implement minimal fix, run tests and sniper before PR
  • Refactor a React component: split large files, extract interfaces, follow SOLID and run self-review checklist
  • Ship a hotfix branch with minimal changes, validation, and urgent merge steps
  • Onboard a contributor by auto-selecting framework guides and enforcing APEX phases

FAQ

Yes. Initialize APEX tracking before any write/edit; the PreToolUse hooks block changes until tracking is created.

What happens if a file exceeds 100 lines?

Stop and split the file immediately. The workflow enforces a hard limit: never exceed 100 lines.

Can I skip the expert self-review?

You may use --skip-elicit mode, but expert self-review (eLicit) is recommended and part of the default (--auto) flow.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
apex skill by fusengine/agents | VeilStrat