kata-new-project
SkillDev toolsUse when the user wants to start, define, import, or initialize a Kata project.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the kata-new-project skill
What this skill tells your AI
The instructions your AI receives, as published by gannonh/kata-symphony in apps/cli/skills/kata-new-project/SKILL.md and read by ahel’s review.
Operating Brief
When this skill is invoked, help the user initialize durable Kata project context.
Start conversationally. If the user has not already provided enough detail, ask: "What do you want to build?" Follow their thread until you can explain the project, why it matters, who it is for, what done looks like, and what is explicitly out of scope.
Before durable writes, run node ./scripts/kata-call.mjs health.check. Then confirm the synthesized project brief with the user, run project.upsert, and write project-scoped artifacts with artifact.write.
This skill captures project-level context only; milestone scope and roadmap creation happen next in kata-new-milestone.
Success Criteria
- The project exists in the backend through
project.upsert. - A project-scoped
project-briefartifact captures what is being built, core value, users, context, constraints, decisions, and open questions. - A project-scoped
requirementsartifact is written when concrete requirement hypotheses emerged. - The user knows the next step is
kata-new-milestone.
Do Not
- Do not create milestones, slices, or tasks.
- Do not write durable state outside the CLI backend contract.
- Do not route to standalone discussion skills.
- Do not skip questioning when the project idea is vague.
Process
- Read
references/workflow.mdbefore taking action. Execute that workflow end-to-end. - Preserve every workflow gate: required checks, user confirmations, durable writes, status updates, and next-step routing.
- Before any backend IO, read
references/runtime-contract.mdand use only the operations listed there. - When the workflow tells you to create or read an artifact, use
references/artifact-contract.mdand the named template files. - If setup or backend readiness is uncertain, read
references/setup.mdbefore proceeding. - Read optional references only when the workflow calls for them or the current step needs them.
Resource Loading
Must read:
- Workflow:
references/workflow.md - Runtime IO contract:
references/runtime-contract.md
Read when needed:
- Setup and health checks:
references/setup.md - Alignment depth:
references/alignment.md - CLI command patterns:
references/cli-runtime.md - Artifact conventions:
references/artifact-contract.md - CLI helper:
scripts/kata-call.mjs - Artifact input helper:
scripts/kata-artifact-input.mjs
Additional references:
- questioning:
references/questioning.md - ui-brand:
references/ui-brand.md
Templates:
- project:
templates/project.md - requirements:
templates/requirements.md
Execution Rules
- If setup or backend state is uncertain, start with
references/setup.md. - Choose alignment depth using
references/alignment.mdinside this workflow. - Follow
references/workflow.mdas the behavioral source for this skill. - Use only operations listed in
references/runtime-contract.mdfor backend IO. - Keep backend specifics in @kata-sh/cli adapters, never in skill logic.
Signals
- GitHub stars
- 63
- Forks
- 6
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Gateway key
kata-new-project- Source
- github.com/gannonh/kata-symphony