A portable, provider-neutral agent pack for the Kiro CLI.
One orchestrator. Eight specialists. Zero lock-in.
A single AI agent juggling planning, coding, reviewing, and deploying loses context, makes sloppy mistakes, and has way too many permissions. You end up babysitting it.
Kodama fixes this by splitting the work across specialists — the scout can only read, the smith can only write with approval, the critic only reviews. Each agent does one job well, with the minimum permissions needed.
You describe what you want. Kodama figures out who does what.
No provider lock-in. Works with any model supported by Kiro CLI.
Implementation tools require approval. Read-only agents stay read-only.
9 purpose-built agents. Clear boundaries, no overlap.
Work is reviewed and validated before completion.
Kodama delegates to purpose-built specialists. Each has clear boundaries and minimal permissions.
Orchestrator
Plans, delegates, reconciles, verifies
Reconnaissance
Read-only codebase exploration
Research
External docs, APIs, libraries
Reasoning
Architecture, debugging, tradeoffs
UI/UX
Components, styling, accessibility
Implementation
Code, tests, refactors
Review
Independent diff review
Infrastructure
CI/CD, Docker, IaC, deploy
Documentation
READMEs, ADRs, changelogs
One command. Requires kiro-cli and python3.
One-liner install
curl -fsSL https://dl.getkodama.dev | bashOr directly from GitHub
curl -fsSL https://raw.githubusercontent.com/amergrgic/kodama/main/install-remote.sh | bashOr clone and run locally
git clone https://github.com/amergrgic/kodama.git
cd kodama && ./setup.shThen start a session:
kiro-cli chat --agent kodamaOr install with --alias to just type kodama:
./setup.sh --aliasSame prompt. Different approach.
Default Kiro agent
Kodama