kalash33/kstack/agent-build

Claude Code and agent skill by kalash33/kstack

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Build AI agents with CrewAI, LangGraph, PydanticAI, or custom implementations. Handles tool creation, prompt engineering, memory setup, and orchestration. Use when: "build an agent", "implement the agent", "create a crew", "langgraph workflow", or after /agent-design.

Install

npx vskill@latest install kalash33/kstack/agent-build

SKILL.md

The same SKILL.md was published earlier as kalash33/kstack/agent-build.

/agent-build — Agent Builder

Turn designs into working agents.

When to Use

  • After /agent-design produces an architecture
  • Building a new agent from scratch
  • Adding agents to an existing system
  • Migrating between frameworks

Supported Frameworks

CrewAI

Best for: Role-based teams with clear task delegation

python
from crewai import Agent, Task, Crew
agent = Agent(
role="Researcher",
goal="Find accurate information",
tools=[search_tool, scrape_tool],
llm=llm
)
LangGraph

Best for: Complex workflows with cycles, branching, human-in-the-loop

python
from langgraph.graph import StateGraph
graph = StateGraph(AgentState)
graph.add_node("research", research_node)
graph.add_node("write", write_node)
graph.add_edge("research", "write")
PydanticAI

Best for: Type-safe agents with structured outputs

python
from pydantic_ai import Agent
agent = Agent(
'openai:gpt-4o',
system_prompt="You are a helpful assistant",
result_type=StructuredOutput
)
Custom (Python/TypeScript)

Best for: Full control, minimal dependencies, production systems

Process

Step 1: Scaffold

Based on the design doc, create:

  • Agent definitions (roles, prompts, tools)
  • Tool implementations
  • Memory/state management
  • Orchestration logic
Step 2: Implement Tools

Every tool follows this pattern:

python
def tool_name(param: str) -> str:
"""Clear description of what this tool does.
Args:
param: What this parameter means
Returns:
What the tool returns
"""
# Implementation
return result
Step 3: Wire Up Memory
  • Short-term: Conversation buffer (last N messages)
  • Long-term: Vector store (ChromaDB, Pinecone, pgvector)
  • Episodic: Key decisions and outcomes
  • Structured: Knowledge graph for relationships
Step 4: Test Locally
bash
# Run the agent with a test prompt
python -m agent.main "test task description"
Step 5: Iterate
  • Check tool call accuracy
  • Verify memory retrieval relevance
  • Test error recovery
  • Measure latency and cost

Chaining

  • Before: /agent-design (provides the architecture)
  • After: /agent-eval (benchmarks the agent)

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/kalash33/kstack/agent-build/badge)](https://verified-skill.com/skills/kalash33/kstack/agent-build)

Something wrong with this skill or its page? Tell us.