Skip to main content
affpapa.org
RU EN
skill Claude Code CX Codex Cursor GitHub Copilot

agentic-engineering — Claude Code skill: what it does, how to install

Establishes protocols for agent-led workflows: breaks work into 15-minute verifiable units, enforces eval-first testing, and routes tasks between Haiku, Sonnet, and Opus.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓ 
npx skills add affaan-m/ECC --skill agentic-engineering -a claude-code
Open on GitHub

What agentic-engineering does

This skill sets up an execution framework where agents write code while developers control boundaries and risk. Work is split using the 15-minute unit rule. Each unit must be independently verifiable, focus on a single dominant risk, and declare a clear done condition. The eval-first loop executes regression checks before code changes, capturing failure signatures before verifying the final delta.

Workloads follow strict model routing to protect budgets. Haiku handles classification and narrow edits, Sonnet takes implementation and refactors, while Opus tackles system architecture and invariant tracking. Escalation happens only when a smaller model hits a visible reasoning wall. Teams track tokens, retries, and wall-clock duration per step.

Install this if your agent loses context on multi-file features or burns quota on trivial chores. For small, single-file scripts, it adds unnecessary overhead.

What it helps with

  • Break complex refactors into 15-minute verifiable steps
  • Capture baseline failure signatures before code generation
  • Route tasks between Haiku, Sonnet, and Opus based on complexity
  • Track token usage, retries, and wall-clock execution time

Example prompts for the agent

  • Break this task into 15-minute units following agentic-engineering and start with evals
  • Plan the authentication refactor using the eval-first approach
  • Route this implementation ticket to the appropriate model tier per agentic-engineering

How to install agentic-engineering

1Skills CLI (Vercel)

npx skills add affaan-m/ECC --skill agentic-engineering -a claude-code

2Manually: put the files in place

~/.claude/skills/agentic-engineering/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/agentic-engineering/

From the same set146

The whole set →
video-editing affaan-m AI-assisted workflow for editing, trimming, and enriching real footage. Connects raw recordings with FFmpeg cuts, Remotion overlays, and ElevenLabs voiceovers. Claude Code CX Codex Cursor +1 6,500installs docker-patterns affaan-m A skill with Dockerfile and Docker Compose patterns: local dev stacks, networking, volumes, container hardening. Claude Code CX Codex Cursor +1 6,300installs frontend-patterns affaan-m A skill for React and Next.js work: component patterns, state, data fetching, and render performance. Use it when building or reviewing UI. Claude Code CX Codex Cursor +1 5,900installs backend-patterns affaan-m A skill with backend patterns for Node.js, Express and Next.js API routes: layering, caching, transactions, error handling. Claude Code CX Codex Cursor +1 5,900installs coding-standards affaan-m Baseline cross-project coding conventions for naming, immutability, and clean code principles. Useful for code reviews and refactoring. Claude Code CX Codex Cursor +1 5,600installs api-design affaan-m Provides REST API design standards covering resource naming, status codes, pagination, and error payloads. Use it when creating or auditing API contracts. Claude Code CX Codex Cursor +1 5,500installs rust-testing affaan-m A skill for writing Rust tests: unit, integration, async, property-based, mocks and coverage. Useful when the project runs on TDD. Claude Code CX Codex Cursor +1 5,500installs laravel-security affaan-m This skill reviews Laravel authentication, Eloquent safety, CSRF, XSS, APIs, and production settings. Use it for security reviews or deployment checks. Claude Code CX Codex Cursor +1 5,400installs

Similar

find-skills vercel-labs Discovers and installs agent skills from skills.sh directly inside your coding assistant. Handy when you need domain-specific workflows for a project. Claude Code CX Codex Cursor +1 4,000,000installs grill-me Matt Pocock grill-me interviews you about a plan or design to expose gaps and sharpen decisions. Use it before implementation starts. Claude Code 1,424,476installs grill-with-docs Matt Pocock This skill interviews you about a plan or design and creates ADRs and a glossary along the way. Use it when decisions need sharper definition. Claude Code 1,219,756installs handoff Matt Pocock Creates a handoff document from the current conversation so another agent can continue the work. Use it when a task moves to a new session or agent. Claude Code 1,038,491installs grilling Matt Pocock Stress-test a plan, decision, or idea through structured rounds of questions. Use it when you need to expose assumptions before acting. Claude Code 979,543installs teach Matt Pocock Turns the current directory into a learning workspace and keeps progress across sessions. Use it when learning needs to support a concrete goal in your project. Claude Code 862,247installs

Questions about agentic-engineering

How do I install agentic-engineering in Claude Code?

Run npx skills add affaan-m/ECC --skill agentic-engineering -a claude-code. The rule file will be placed in ~/.claude/skills/agentic-engineering/SKILL.md.

What is the purpose of agentic-engineering?

It structures autonomous development by requiring agents to define completion criteria, run baseline tests before coding, and split tasks into short increments.

When does the skill escalate to a higher model tier?

Escalation to Sonnet or Opus occurs only when a lower tier fails because of a clear reasoning gap, preventing wasted spend on simple classification or edits.

Source text

SKILL.md · MIT 
---
name: agentic-engineering
description: Operate as an agentic engineer using eval-first execution, decomposition, and cost-aware model routing. Use when planning or executing engineering work that agents will carry out end to end.
metadata:
  origin: ECC
---

# Agentic Engineering

Use this skill for engineering workflows where AI agents perform most implementation work and humans enforce quality and risk controls.

## Operating Principles

1. Define completion criteria before execution.
2. Decompose work into agent-sized units.
3. Route model tiers by task complexity.
4. Measure with evals and regression checks.

## Eval-First Loop

1. Define capability eval and regression eval.
2. Run baseline and capture failure signatures.
3. Execute implementation.
4. Re-run evals and compare deltas.

## Task Decomposition

Apply the 15-minute unit rule:
- each unit should be independently verifiable
- each unit should have a single dominant risk
- each unit should expose a clear done condition

## Model Routing

- Haiku: classification, boilerplate transforms, narrow edits
- Sonnet: implementation and refactors
- Opus: architecture, root-cause analysis, multi-file invariants

## Session Strategy

- Continue session for closely-coupled units.
- Start fresh session after major phase transitions.
- Compact after milestone completion, not during active debugging.

## Review Focus for AI-Generated Code

Prioritize:
- invariants and edge cases
- error boundaries
- security and auth assumptions
- hidden coupling and rollout risk

Do not waste review cycles on style-only disagreements when automated format/lint already enforce style.

## Cost Discipline

Track per task:
- model
- token estimate
- retries
- wall-clock time
- success/failure

Escalate model tier only when lower tier fails with a clear reasoning gap.

© affaan-m · MIT · affaan-m/ECC

See also

Project sponsors

Companies that keep this analytics open
Gold sponsor
Nutra.Club

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor