Orca is the place for working with a fleet of agents (e.g. Grok Bot, Claude Code). Run any coding agent with your own subscription. Available on desktop, mobile and remote runtime.
TypeScript adeagent-ideai-agentsautomationbotclaude-codecodexghosttygrok-bothermes-agentmobile-appmuseopenclawopencodeorchestrationpiruntimeterminalworktreesyc-backed
// Share grab extraction contracts across main/preload/renderer, including redaction and budgets.
/** Page-level metadata captured at selection time. */
export type BrowserGrabPageContext = {
sanitizedUrl: string
title: string
viewportWidth: number
viewportHeight: number
scrollX: number
scrollY: number
devicePixelRatio: number
capturedAt: string
}
/** Accessibility metadata for the selected element. */
export type BrowserGrabAccessibility = {
role: string | null
accessibleName: string | null
ariaLabel: string | null
ariaLabelledBy: string | null
}
/** Curated subset of computed styles. */
export type BrowserGrabComputedStyles = {
display: string
position: string
width: string
height: string
margin: string
padding: string
color: string
backgroundColor: string
border: string
borderRadius: string
fontFamily: string
fontSize: string
fontWeight: string
lineHeight: string
textAlign: string
zIndex: string
}
/** Viewport-relative or page-relative rectangle in CSS pixels. */
export type BrowserGrabRect = {
x: number
y: number
width: number
height: number
}
/** The selected element's extracted data. */
export type BrowserGrabTarget = {
tagName: string
selector: string
elementPath?: string
fullPath?: string
cssClasses?: string
nearbyElements?: string[]
selectedText?: string | null
isFixed?: boolean
reactComponents?: string | null
sourceFile?: string | null
textSnippet: string
htmlSnippet: string
attributes: Record<string, string>
accessibility: BrowserGrabAccessibility
rectViewport: BrowserGrabRect
rectPage: BrowserGrabRect
computedStyles: BrowserGrabComputedStyles
}
/** Screenshot attachment — always PNG, either a data URL or a temp file path. */
export type BrowserGrabScreenshot = {
mimeType: 'image/png'
dataUrl: string
width: number
height: number
}
/** The full payload extracted from a browser grab selection. */
export type BrowserGrabPayload = {
page: BrowserGrabPageContext
target: BrowserGrabTarget
nearbyText: string[]
ancestorPath: string[]
screenshot: BrowserGrabScreenshot | null
}
/** Persisted annotation payloads keep DOM context but drop transient screenshots. */
export type BrowserAnnotationPayload = Omit<BrowserGrabPayload, 'screenshot'> & {
screenshot: null
}
export type GrabIntent = 'copy' | 'annotate'
/** Why a grab operation was cancelled before the user selected an element. */
export type BrowserGrabCancelReason = 'user' | 'tab-inactive' | 'navigation' | 'evicted' | 'timeout'
/** Discriminated union for the result of a single grab operation. */
export type BrowserGrabResult =
| { opId: string; kind: 'selected'; payload: BrowserGrabPayload }
| { opId: string; kind: 'context-selected'; payload: BrowserGrabPayload }
| { opId: string; kind: 'cancelled'; reason: BrowserGrabCancelReason }
| { opId: string; kind: 'error'; reason: string }
export type BrowserSetGrabModeArgs = {
browserPageId: string
enabled: boolean
}
/** Why a grab IPC call was rejected before the operation could start. */
export type BrowserGrabRejectReason =
| 'not-ready'
| 'not-authorized'
| 'already-active'
| 'injection-failed'
export type BrowserSetGrabModeResult = { ok: true } | { ok: false; reason: BrowserGrabRejectReason }
export type BrowserAwaitGrabSelectionArgs = {
browserPageId: string
opId: string
}
export type BrowserCancelGrabArgs = {
browserPageId: string
}
export type BrowserCaptureSelectionScreenshotArgs = {
browserPageId: string
rect: BrowserGrabRect
}
export type BrowserCaptureSelectionScreenshotResult =
| { ok: true; screenshot: BrowserGrabScreenshot }
| { ok: false; reason: string }
export type BrowserExtractHoverArgs = {
browserPageId: string
}
export type BrowserExtractHoverResult =
| { ok: true; payload: BrowserGrabPayload }
| { ok: false; reason: string }
export type BrowserAnnotationIntent = 'fix' | 'change' | 'question' | 'approve'
export type BrowserAnnotationPriority = 'blocking' | 'important' | 'suggestion'
export type BrowserPageAnnotation = {
id: string
browserPageId: string
comment: string
intent: BrowserAnnotationIntent
priority: BrowserAnnotationPriority
createdAt: string
payload: BrowserAnnotationPayload
/** Pasted images, saved on the worktree's host like a terminal image paste. */
imagePaths?: string[]
}
// Payload budgets are enforced in guest and main.
export const GRAB_BUDGET = {
textSnippetMaxLength: 200,
nearbyTextEntryMaxLength: 200,
nearbyTextMaxEntries: 10,
htmlSnippetMaxLength: 4096,
ancestorPathMaxEntries: 10,
nearbyElementsMaxEntries: 6,
nearbyElementMaxLength: 160,
selectorMaxLength: 700,
pathMaxLength: 900,
cssClassesMaxLength: 500,
selectedTextMaxLength: 500,
sourceFileMaxLength: 500,
reactComponentsMaxLength: 500,
annotationCommentMaxLength: 2000,
annotationsMaxPerPage: 20,
/** Hard byte budget for screenshot PNG data URL before we omit the screenshot. */
screenshotMaxBytes: 2 * 1024 * 1024
} as const
export function isBrowserAnnotationPageFull(annotationCount: number): boolean {
return annotationCount >= GRAB_BUDGET.annotationsMaxPerPage
}
/** Only these attribute names are included in the payload by default. */
export const GRAB_SAFE_ATTRIBUTE_NAMES = new Set([
'id',
'class',
'name',
'type',
'role',
'href',
'src',
'alt',
'title',
'placeholder',
'for',
'action',
'method'
])
/** Attribute names matching aria-* are always included. */
export function isAriaAttribute(name: string): boolean {
return name.startsWith('aria-')
}
/** Redact OAuth/credential-shaped values, not broad words like code/state that match ordinary CSS classes and degrade extraction. */
export const GRAB_SECRET_PATTERNS = [
'access_token',
'auth_token',
'api_key',
'apikey',
'client_secret',
'oauth_state',
'x-amz-',
'session_id',
'sessionid',
'csrf',
'secret',
'password',
'passwd'
]