ak-init

A project starter for building a new AI agent with the Agent Kernel library. It helps choose an agent framework, tools, deployment target, and project structure.

In plain words
What is it for?
Use it to start agents such as customer-support bots, code-review assistants, or data-analysis agents, with optional custom tools and multiple specialized agents.
Why use it?
It removes the need to decide the basic architecture and create all project files by hand. It also guides you through the requirements your agent needs.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/yaalalabs/agent-kernel/ak-init
Any agent
npx skills add yaalalabs/agent-kernel --skill ak-init
Clone the repo
git clone --depth 1 https://github.com/yaalalabs/agent-kernel

Made for: Claude Code, Codex.

Per session 58 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,848 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00058 $0.02848
Opus 5 $0.00029 $0.01424
Sonnet 5 $0.00012 $0.00570
Haiku 4.5 $0.00006 $0.00285

Measured 2d ago against content hash ee6fcde42fbd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ak-init scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

ak-py/src/agentkernel/skills/ak-init/SKILL.md · 395 lines

How it starts

The opening of the file, as written. The whole thing — 395 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Scaffold an Agent Kernel Project

Use this skill to create a new AI agent project powered by Agent Kernel.

Instructions for the Agent

When the user wants to create a new agent project, follow this interactive workflow:

Step 1: Gather Requirements

Ask the user the following questions (adapt based on context):

  1. Agent framework: Which agent framework would you like to use?

    • OpenAI Agents SDK (recommended for most use cases — best tool support, handoffs between agents)
    • CrewAI (multi-agent collaboration with roles and tasks)
    • LangGraph (complex workflow graphs with state management)
    • Google ADK (Google's Agent Development Kit)
    • Smolagents (lightweight agent framework with managed-agent routing)
    • Pydantic AI (provider-agnostic — native OpenAI/Anthropic/Google/Bedrock/… support with FallbackModel failover)
  2. Agent purpose: What should your agent(s) do? (e.g., "customer support bot", "code review assistant", "data analysis agent")

  3. Tools: Does your agent need any custom tools? (e.g., "fetch weather data", "query a database", "search the web")

  4. Multi-agent: Do you need multiple specialized agents with a triage/routing agent?

  5. Deployment mode: How will you run the agent?

    • CLI (interactive terminal — great for development and testing)
    • REST API (FastAPI server — for web apps, webhooks, integrations)
    • AWS Lambda (serverless on AWS)
    • AWS ECS/Fargate (containerized on AWS)
    • Azure Functions (serverless on Azure)
    • Azure Container Apps (containerized on Azure)
    • GCP Cloud Run Serverless (scale-to-zero on GCP)
    • GCP Cloud Run Containerized (always-on on GCP)
    • Docker (generic container, runs anywhere)
  6. Session persistence: How should conversation state be stored?

    • In-memory (default, no persistence — fine for CLI and development)
    • Redis (recommended for production — works with all deployment targets)
    • DynamoDB (AWS-native, recommended for AWS serverless)
    • Cosmos DB (Azure-native, recommended for Azure serverless)
    • Firestore (GCP-native, recommended for GCP Cloud Run)

Read the full file on GitHub · 395 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 395 lines · 58 tokens per session scan A ee6fcde42fbd

Subscribe to this mod's changes

ak-init is a skill published in the GitHub repository yaalalabs/agent-kernel (145 stars, last pushed 5d ago), licensed Apache-2.0. It adds 58 tokens to every session and 2,848 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.