Agentic Engineering
- 1 installs
- 3 repo stars
- Updated June 27, 2026
- agentmatters/mullai-bot
agentic-engineering is a Claude Code skill that defines an eval-first, decomposition-driven method for running engineering work through AI agents with human quality control.
About
A Claude Code skill that defines an operating method for agentic engineering, where AI agents do most implementation and humans enforce quality. It covers eval-first loops, task decomposition, cost-aware model routing, session strategy, and review focus for AI-generated code. A developer uses it to structure and control multi-step agent work.
- Operating principles for eval-first, agent-driven engineering
- Model routing guidance across Haiku, Sonnet, and Opus by task complexity
- Task decomposition into independently verifiable 15-minute units
Agentic Engineering by the numbers
- 1 all-time installs (skills.sh)
- Ranked #14,098 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Jul 7, 2026 (Skillselion catalog sync)
agentic-engineering capabilities & compatibility
- Capabilities
- agent orchestration · task decomposition · code review
- Use cases
- orchestration · code review · planning
- Pricing
- Free
What agentic-engineering says it does
Use this skill for engineering workflows where AI agents perform most implementation work and humans enforce quality and
Escalate model tier only when lower tier fails with a clear reasoning gap.
npx skills add https://github.com/agentmatters/mullai-bot --skill agentic-engineeringAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1 |
|---|---|
| repo stars | ★ 3 |
| Last updated | June 27, 2026 |
| Repository | agentmatters/mullai-bot ↗ |
What it does
Run engineering work as agent-driven execution with evals, decomposition, and cost-aware model routing.
Who is it for?
Structuring engineering work where agents implement and humans enforce quality and risk controls.
Skip if: Concrete implementation of a specific framework or API.
When should I use this skill?
Operating as an agentic engineer using eval-first execution, decomposition, and cost-aware model routing.
What you get
A repeatable operating loop for agent-driven engineering with evals, decomposition, and cost discipline.
- Eval-first execution loop
- Task decomposition plan
- Model routing decisions
By the numbers
- Defines a 15-minute unit rule for task decomposition
- Lists a 3-tier model routing scheme (Haiku, Sonnet, Opus)
Files
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.
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10FAQ
What is the model routing guidance?
Haiku for classification and boilerplate, Sonnet for implementation and refactors, Opus for architecture and root-cause analysis.
What is the task decomposition rule?
The 15-minute unit rule: each unit is independently verifiable, has a single dominant risk, and a clear done condition.