.agents/skills/oat-project-new/SKILL.md
Use when starting a spec-driven OAT project from scratch. Scaffolds a new project under PROJECTS_ROOT and sets it active.
npx skillsauth add tkstang/open-agent-toolkit oat-project-newInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Create a new OAT project directory, scaffold standard artifacts from .oat/templates/, and set activeProject in local config.
Print a phase banner once at start using horizontal separators, e.g.:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ OAT ▸ NEW PROJECT ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Before multi-step work, print step indicators, e.g.:
[1/3] Validating project name…[2/3] Scaffolding project artifacts…[3/3] Refreshing dashboard…Resolve {PROJECTS_ROOT} (same order as other OAT skills):
PROJECTS_ROOT="${OAT_PROJECTS_ROOT:-$(oat config get projects.root 2>/dev/null || echo \".oat/projects/shared\")}"
PROJECTS_ROOT="${PROJECTS_ROOT%/}"
If not provided in $ARGUMENTS, ask the user for {project-name} (slug format: alphanumeric/dash/underscore only).
Use the CLI scaffolder:
oat project new "{project-name}" --mode spec-driven
Optional flags:
--force (non-destructive; only fills missing files/dirs, does not overwrite)--no-set-active--no-dashboardConfirm to the user:
{PROJECTS_ROOT}/{project-name}.oat/config.local.json (activeProject).oat/state.md (if enabled)Then explicitly instruct the user to run discovery next:
oat-project-discover{PROJECTS_ROOT}/{project-name}/ exists.oat/templates/*.md)activeProject in .oat/config.local.json points at the project path.oat/state.md is refreshed (unless disabled)documentation
Use when OAT implementation changes and repository reference docs must be synchronized. Updates .oat/repo/reference to match current behavior.
business
Merge multiple analysis artifacts into a single coherent report with provenance tracking. Reads existing artifacts from /deep-research, /analyze, and /compare.
testing
Use when the user questions or suspects an agent claim is wrong. Adversarially gathers evidence to verify or refute the claim using the best sources available in the current environment.
tools
Use when prioritizing backlog work or evaluating a roadmap. Produces value-effort ratings, dependency mapping, and execution recommendations.