agent-skills

Set up an agent-friendly local dev workflow — a bin/ task-runner (inspired by GitHub's scripts-to-rule-them-all) over a shared Postgres container that gives every git worktree its own isolated database and ports, plus a dedicated test database so tests never clobber dev data. Use this whenever a project needs worktree-isolated local development, when setting up for AI agent orchestrators (Conductor and similar) that spin up a worktree per session and register setup/run/cleanup commands, when multiple copies of a backend must run concurrently on one machine, when replacing a docker-compose-for-local-postgres setup, when deciding whether auxiliary dev services (a validator container, a local OIDC IdP) replicate per worktree or run shared, when merged agent worktrees pile up and need sweeping, or when tests keep wiping local dev/demo data. Triggers on "bin/ scripts", "worktree isolation", "per-worktree database/port", "scripts to rule them all", "agent dev environment", "setup/run/cleanup scripts", "shared auxil

Stand up the pre-merge CI quality gates a repo runs before code lands on develop — tool provisioning (asdf + .tool-versions + cache), lockfile-frozen installs, path-filtered GitHub Actions, and the lint / format:check / typecheck / test gate set with Jarvus's standard linters (oxlint + oxfmt for TypeScript, ruff for Python, tofu fmt/validate for IaC). Use when setting up CI for a new repo, adding a lint/format/type-check/test gate to an existing one, wiring GitHub Actions for code quality, choosing or standardizing linters/formatters, fixing CI that re-installs tools slowly, or noticing a repo runs tests but no linter. Scope stops at merge — Release-PR automation belongs to release-flow; build/publish/deploy to per-stack build + deploy skills. Triggers: "set up CI", "add linting/CI", "GitHub Actions checks", "oxlint", "oxfmt", "ruff", "format check", "type-check in CI", "quality gate", "lint isn't running in CI", "asdf in CI".

Backend development using Fastify + TypeScript on Bun. Use when creating new backend APIs, adding routes, implementing services, working with plugins, or configuring environment variables.

Cut a release in a repo that uses the Jarvus develop→main Release-PR automation (the JarvusInnovations/infra-components release-prepare/validate/publish GitHub Actions). Use this whenever you're shipping merged work to production, drafting or editing a "Release: vX.Y.Z" PR, deciding a version bump, writing release notes/changelog, or the user says "ship it", "cut a release", "publish", "do the release", "merge the release PR", or "release notes". Also use it the moment you notice a repo has `.github/workflows/release-prepare.yml` (or release-publish/release-validate) or an open PR titled "Release: v*" — that's the signal this workflow is in effect. Covers: pushing develop to open the Release PR, pulling the bot-generated changelog, sorting commits into Improvements vs Technical, recommending semver bump, and merging to publish.

Jarvus house conventions for writing, testing, and linting dbt projects — model layering and grain, what each stage may/may not do, the no-inner-joins and semantic-alias rules, the generic + quality-model + unit-test patterns, the sqlfluff config, and a credential-free multi-tenant CI gate. Use when building or reviewing dbt models, deciding materialization or test coverage, setting up dbt linting/CI, or when "dbt", "sqlfluff", "staging/intermediate/marts", "dbt test", or "TIDES" come up. This is the opinionated house layer ON TOP OF dbt-labs' first-party dbt skills (which cover the mechanics) — see "Relationship to other skills".

Mobile app development using Flutter + Riverpod + go_router. Use when creating Flutter apps, adding screens, managing state with Riverpod, implementing routing, building offline-first features, or setting up storybook component development. Also use when the user mentions Flutter, Dart, mobile apps, or cross-platform development.

Frontend development using Bun + Vite + React + Tailwind CSS + React Router v7. Use when creating new frontend projects, adding UI components, implementing routing, styling with Tailwind, or working with the React frontend stack. shadcn/ui is an optional component-library layer (see references/shadcn.md).

Packaging-and-distribution recipe for baking an AXI CLI (axi-sdk-js) into a Claude Code skill — bundling it under the skill's scripts/ as a self-contained committed .mjs shipped via `npx skills add`. This is the build/ship companion to the upstream `axi` skill: `axi` covers what good agent-facing CLI output looks like (TOON, schemas, content-first); this covers how to ship that CLI *inside a skill*. Use whenever you're turning an AXI tool into an installable skill, or working on any of its moving parts — the esbuild → committed .mjs build, the bash shim, the SessionStart hook (install/uninstall/status, project vs global scope), generating SKILL.md from a single-source command reference, the CI drift gate, or splitting a lean every-session "home" view from an on-demand live "dashboard". Triggers: "bundle my CLI in a skill", "AXI in a skill", "ship this CLI via npx skills", "self-contained skill bundle", "add a session-start hook to my skill", "generate SKILL.md from the CLI", or noticing a repo with `scripts/b