Agentic Feature Design
Designing features for the "Action Era" that are AI-accessible by default
Best use case
Agentic Feature Design is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Designing features for the "Action Era" that are AI-accessible by default
Teams using Agentic Feature Design should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/agentic-feature-design/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How Agentic Feature Design Compares
| Feature / Agent | Agentic Feature Design | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Designing features for the "Action Era" that are AI-accessible by default
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Agentic Feature Design
Features in LivestockAI must now be designed for **dual consumption**: Humans (UI) and Agents (MCP/API).
## 1. The "Headless First" Rule
Every feature must be fully functional via Server Functions _before_ any UI is built.
**Test:** Can an agent complete the entire user story using only `bun run check-feature-x.ts` (a script calling server functions)?
If yes -> **Agent Ready.**
If no (logic lives in React components) -> **Refactor immediately.**
## 2. Agent-Ready Server Functions
Server functions are the tools we give to agents. Document them accordingly.
```typescript
export const createBatchFn = createServerFn({ method: 'POST' })
.inputValidator(batchSchema)
.handler(async ({ data }) => {
/* ... */
})
/**
* @description Creates a new livestock batch.
* @workflow
* 1. Check `getFarmFacilities` for space.
* 2. `createBatchFn`
* 3. Log initial `feed_record` if applicable.
*/
```
## 3. The "Intention" Pattern
Agents operate on _Intent_, not just data.
Instead of generic CRUD (`updateBatch`), expose semantic actions (`graduateBatch`, `quarantineBatch`).
```typescript
// ❌ Generic
updateBatch(id, { status: 'sold' })
// ✅ Semantic (Agent Friendly)
markBatchAsSold(id, { date, price, customer })
```
Semantic actions allow agents to:
1. Understand the _consequences_ of the action.
2. Perform specialized validation (e.g. "Can't sell a quarantined batch").
## 4. Approval Loops (Human-in-the-Loop)
Agents need a standard way to ask for permission for high-stakes actions (e.g., ordering feed, selling stock).
**The `ApprovalRequest` Entity:**
- `agentId`: Who is asking?
- `action`: JSON payload of the server function to call.
- `summary`: Human-readable explanation ("I want to order 50 bags of Starter Feed").
- `status`: PENDING | APPROVED | REJECTED
## 5. Metadata for Context
All major entities (`Batches`, `Sales`) should have an `ai_metadata` JSONB column.
Agents use this to store reasoning ("Predicted harvest date change due to low feed intake").
_Do not show this raw JSON to users, but use it to power UI "Insights"._
## Related Skills
- `three-layer-architecture` - Where the logic lives
- `feature-structure` - How to organize the filesRelated Skills
feature
Creates a new feature module with minimal viable structure. Use when bootstrapping a new feature from scratch, scaffolding the Tuist module, Container, Feature entry point, DeepLinkHandler, and initial screen with placeholder Text view. Includes all unit tests, mocks, stubs, and app integration. For adding domain/data layers afterward, use /datasource, /repository, /usecase. For enhancing views, use /view, /viewmodel, /navigator.
domain-driven-design
Plan and route Domain-Driven Design work from strategic modeling to tactical implementation and evented architecture patterns.
data-designer
Generate high-quality synthetic datasets using statistical samplers and Claude's native LLM capabilities. Use when users ask to create synthetic data, generate datasets, create fake/mock data, generate test data, training data, or any data generation task. Supports CSV, JSON, JSONL, Parquet output. Adapted from NVIDIA NeMo DataDesigner (Apache 2.0).
analytics-design
Design data analysis from purpose clarification to visualization. Use when analyzing data, exploring BigQuery schemas, building queries, or creating Looker Studio reports.
agenticmail
🎀 AgenticMail — Full email, SMS, storage & multi-agent coordination for AI agents. 63 tools.
agentic-issue-assistant
Install common docs/backlog skeleton plus an AGENTS template, and wrap issue/finalization operations for an agentic workflow.
agentic-chat
AI assistant for creating clear, actionable task descriptions for GitHub Copilot agents
---name: aav-vector-design-agent
description: AI-powered adeno-associated virus (AAV) vector design for gene therapy including capsid engineering, promoter selection, and tropism optimization.
Data Engineering Data Driven Feature
World-class data science skill for statistical modeling, experimentation, causal inference, and advanced analytics. Expertise in Python (NumPy, Pandas, Scikit-learn), R, SQL, statistical methods, A/B testing, time series, and business intelligence. Includes experiment design, feature engineering, model evaluation, and stakeholder communication.
Schema Design
Migration-ready database schema design with normalization and indexing strategies
implement-feature
Implementa feature nel sistema di fatturazione italiana validando contro normativa fiscale. Usa per aggiungere calcoli IVA, ritenuta d'acconto, split payment, imposta di bollo, gestione fatture PA, regime forfettario, numerazione progressiva, note di credito, o qualsiasi logica che deve rispettare DPR 633/72, DPR 600/73, DPR 642/72. NON usare per bug fix tecnici, refactoring, o modifiche UI senza impatto fiscale.
fullstack-feature
Load PROACTIVELY when task involves building a complete feature across multiple layers. Use when user says "build a feature", "add user profiles", "create a dashboard", or any request spanning database, API, UI, and tests. Orchestrates multi-agent work sequentially: schema and migrations, API endpoints, UI components, tests, and review. The runtime engine handles WRFC chains automatically via <gv> directives. Handles dependency ordering and cross-layer type sharing.