Skill: Build and Publish Kit

Build a Bottlerocket kit (core-kit or kernel-kit) and publish it to a local OCI registry for development testing and validation.

Roles

You (reading this file) are the orchestrator.

Role	Reads	Does
Orchestrator (you)	SKILL.md, next-step.py output	Runs state machine, spawns subagents, writes outputs
State machine	progress.json, workspace files	Decides next action, validates gates
Subagent	Phase file (e.g., SETUP.md)	Executes phase instructions

⚠️ You do NOT read files in phases/ — pass them to subagents via context_files. Subagents read their phase file and execute it.

Orchestrator Loop

python

workspace = f"planning/build-kit-{kit_name}"
bash(f"mkdir -p {workspace}", on_error="raise")

input_data = {"kit_name": kit_name, "arch": arch}
write("create", f"{workspace}/input.json", file_text=json.dumps(input_data))

while True:
    result = bash(f"python3 skills/build-kit-locally/next-step.py {workspace}", on_error="raise")
    action = json.loads(result)
    
    if action["type"] == "done":
        verify = fs_read("Line", f"{workspace}/04-verify.md", 1, 100)
        break
    
    if action["type"] == "gate_failed":
        log(f"Gate failed: {action['reason']}")
        break
    
    if action["type"] == "spawn":
        r = spawn(
            action["prompt"],
            context_files=action["context_files"],
            context_data=action["context_data"],
            allow_tools=True
        )
        write("create", f"{workspace}/{action['output_file']}", file_text=r.response)

Handling Exceptions

The state machine handles the happy path. When things go wrong, exercise judgment:

Exception	Response
Spawn times out	Assess: retry with longer timeout? Report partial progress?
Spawn returns error	Report failure to state machine, let it track retries
Empty/invalid response	Treat as failure, report to state machine

Don't silently advance past failures. Either retry, fail explicitly, or document gaps.

Anti-Patterns

❌ Don't	✅ Do
Read phase files yourself	Pass phase files via context_files to subagents
Decide what phase is next	State machine decides via next-step.py
Skip gates "because it looks done"	Always validate gates
Store state in your memory	State lives in progress.json
Silently advance past failures	Retry, fail, or document gaps

Phases

SETUP: Ensure local registry is running and configure Infra.toml
BUILD: Build the kit for specified architecture
PUBLISH: Publish built kit to local OCI registry
VERIFY: Verify kit was successfully published

Inputs

Create {{workspace}}/input.json with:

kit_name: Name of the kit (e.g., "bottlerocket-core-kit")
arch: Architecture to build (default: "x86_64", or "aarch64")

Outputs

{{workspace}}/01-setup.md: Registry and Infra.toml configuration status
{{workspace}}/02-build.md: Build artifacts and status
{{workspace}}/03-publish.md: Published tags and registry info
{{workspace}}/04-verify.md: Verification results and next steps

When to Use

Making changes to kit packages and testing them in variants
Iterative development on kits
Before building a variant image that depends on kit changes

Prerequisites

Docker installed and running
Kit repository cloned in kits/ directory
FOREST_ROOT environment variable set

Common Issues

Registry not running:

Error: connection refused

Solution: The SETUP phase will start the registry automatically

Infra.toml not configured:

Error: vendor 'local' not found

Solution: The SETUP phase creates Infra.toml if missing

Docker permission denied: Solution: Ensure user is in docker group and Docker daemon is running

Next Steps

After the skill completes:

Update variant's Twoliter.toml to reference the new kit version
Run make update in the variant repo
Build the variant with cargo make

Search AI Tools

build-kit-locally

Install this agent skill to your Project

SKILL.md