agent-native

How to add or edit localized UI copy in agent-native apps. Use when adding, removing, or changing user-visible interface text, prompts, toasts, labels, empty states, or date/number/list formatting. Apps opt in to additional locales from `agent-native.config.ts`.

Server-side and client-side LaunchDarkly flag evaluation (project key `agent-native`). Use when reading a LaunchDarkly flag from an action, plugin, or component, or when wiring `LAUNCHDARKLY_SDK_KEY`. Distinct from the framework's own `feature-flags` system.

Run or extend the Desktop Code Agents Multi-Frontier workflow. Use when coordinating Codex and Claude Code subscriptions, collaboration phases, checkpoints, recovery, usage meters, or helper admission.

Use when asking what to do next after a feature pass while avoiding bloat and checking for unfinished MVP work.

Use when adding or reviewing UI navigation so internal routes preserve normal SPA behavior plus Cmd/Ctrl-click and middle-click new-tab semantics.

Create a fresh git branch. Use for an explicit /new-branch request or when a task-owned worktree needs a branch to ship detached work. Keep platform-assigned Builder.io and Fusion branches in place.

Agent observability, evals, feedback, and experiments. Use when adding observability dashboards, configuring trace capture, setting up evals, creating A/B experiments, or collecting user feedback on agent responses.

How to register user-facing setup steps (API keys, OAuth, connecting third-party services) for the sidebar setup checklist. Use when adding a feature that needs initial user configuration.

Keep apps and templates loading fast. Read when adding a data model, a list/read action, a page or sidebar that loads data, or when something loads slowly, or when adding a dependency to the deployed server bundle. Covers column projection, indexing hot-path queries, avoiding N+1 and round-trip waterfalls, cheap polling, not recomputing on every read, cold-start artifact size, and loading placeholders that hold still.

How to keep template code PostgreSQL-specific and hosting-agnostic. Use when defining schemas, writing raw SQL, or creating server routes.

Continue a coding session on a paired always-on computer. Use when work must survive closing the primary laptop, needs remote computer-use testing, or should run on a scheduled/self-hosted execution host.

Autonomous multi-app QA sweep that drives template apps with Playwright MCP. Use only when the user explicitly runs /qa or asks for an end-to-end QA sweep. Do not load it for ordinary feature work, single-page checks, or one-off bug fixes.

Multi-user collaborative editing with Yjs CRDT, SSE fast-path transport, and granular server-side merge. Use when adding real-time collaborative editing to a template, debugging sync issues, or understanding how the agent and humans edit documents simultaneously.

Decide whether a page needs external changes before refresh, then use the opt-in shared SSE and polling transport safely.

The legacy manage-jobs compatibility surface, job files under jobs/, and cron scheduler internals. Use when maintaining manage-jobs integrations or debugging the scheduler.

How the agent must perform writes so they actually persist under the hosted foreground run budget and long-running background handoffs. Use whenever you create, update, delete, or batch-write app data — especially "do this for many items" loops, or any task where the user expects N things to end up saved.

What to say while long work is still running, and what a legal mid-task stop looks like. Use during any run over a few minutes — migrations, sweeps, babysit loops, automations, deploys — and at the moment you are tempted to stop and ask, report a blocker, or answer "how's it going?".

Sweep recent Slack, GitHub issue, Sentry, first-party Agent-Native Analytics error issues, and explicitly linked tracker feedback: first answer reporters, then fix verified bugs and actionable objective UI defects at the owning boundary, require human signoff for subjective UI changes, build features the invoking user endorsed with an :upvote:, and recap every disposition. Use for scheduled or manual sweeps.

Review recent BuilderIO/agent-native human pull requests, approve eligible internal PRs, and merge changes that are ready by Steve's bar. Use for scheduled or manual PR sweeps; flag external non-bug work and unresolved major product or UX decisions to Steve.

ChiliPiper-style pre-booking forms that route prospects to the right event type based on their answers.