feat(pi): Babysit foundations — shared PR/push extraction, five GitHub tools, sandbox lifetime

apps/docs/content/docs/en/integrations/github.mdx

@@ -59,10 +59,12 @@ Fetch PR details including diff and files changed
 |   ↳ `label` | string | Branch label \(owner:branch\) |
 |   ↳ `ref` | string | Branch name |
 |   ↳ `sha` | string | Commit SHA |
+|   ↳ `repo_full_name` | string | Full name \(owner/repo\) of the branch's repository |
 | `base` | object | Branch reference info |
 |   ↳ `label` | string | Branch label \(owner:branch\) |
 |   ↳ `ref` | string | Branch name |
 |   ↳ `sha` | string | Commit SHA |
+|   ↳ `repo_full_name` | string | Full name \(owner/repo\) of the branch's repository |
 | `id` | number | Pull request ID |
 | `number` | number | Pull request number |
 | `title` | string | PR title |

apps/sim/app/workspace/[workspaceId]/w/[workflowId]/components/panel/components/editor/editor.tsx

@@ -650,7 +650,16 @@ export function Editor() {
                                 : undefined
                             }
                           />
-                          {showDivider && <FieldDivider subblockMarker />}
+                          {showDivider && (
+                            <FieldDivider
+                              subblockMarker
+                              className={
+                                regularSubBlocks[index + 1]?.hideDividerBefore
+                                  ? '[&>div]:invisible'
+                                  : undefined
+                              }
+                            />
+                          )}
                         </div>
                       )
                     })}
@@ -707,7 +716,14 @@ export function Editor() {
                             }
                           />
                           {index < advancedOnlySubBlocks.length - 1 && (
-                            <FieldDivider subblockMarker />
+                            <FieldDivider
+                              subblockMarker
+                              className={
+                                advancedOnlySubBlocks[index + 1]?.hideDividerBefore
+                                  ? '[&>div]:invisible'
+                                  : undefined
+                              }
+                            />
                           )}
                         </div>
                       )

apps/sim/app/workspace/[workspaceId]/w/components/preview/components/preview-editor/preview-editor.tsx

@@ -1482,7 +1482,16 @@ function PreviewEditorContent({
                         subBlockValues={subBlockValues}
                         disabled={true}
                       />
-                      {index < visibleSubBlocks.length - 1 && <FieldDivider subblockMarker />}
+                      {index < visibleSubBlocks.length - 1 && (
+                        <FieldDivider
+                          subblockMarker
+                          className={
+                            visibleSubBlocks[index + 1]?.hideDividerBefore
+                              ? '[&>div]:invisible'
+                              : undefined
+                          }
+                        />
+                      )}
                     </div>
                   ))}
                 </div>

apps/sim/blocks/blocks/pi.test.ts

@@ -7,6 +7,14 @@ import { describe, expect, it, vi } from 'vitest'
 // deliberately does not import it — no block imports from `@/executor`. This test is what ties the
 // two copies together, so adding a provider to one and not the other fails here.
 vi.mock('@/lib/api-key/byok', () => ({ getBYOKKey: vi.fn(), getApiKeyWithBYOK: vi.fn() }))
+vi.mock('@/lib/core/config/env', async (importOriginal) => {
+  const original = await importOriginal<typeof import('@/lib/core/config/env')>()
+  return {
+    ...original,
+    getEnv: vi.fn((key: string) => (key === 'NEXT_PUBLIC_E2B_ENABLED' ? 'true' : undefined)),
+    isTruthy: vi.fn((value: unknown) => value === 'true'),
+  }
+})
 
 import { evaluateSubBlockCondition } from '@/lib/workflows/subblocks/visibility'
 import { PiBlock } from '@/blocks/blocks/pi'
@@ -75,3 +83,97 @@ describe('Pi block search fields', () => {
     expect(PiBlock.inputs.searchApiKey).toBeDefined()
   })
 })
+
+describe('Pi Create PR Babysit surface', () => {
+  it('offers exactly Create PR, Review Code, and Local Dev as top-level modes', () => {
+    const mode = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'mode')
+    const options =
+      typeof mode?.options === 'function'
+        ? mode.options()
+        : (mode?.options as Array<{ id: string }> | undefined)
+
+    expect(options?.map(({ id }) => id)).toEqual(['cloud', 'cloud_review', 'local'])
+  })
+
+  it('declares the toggle, required reviewer mentions, advanced rounds, and result outputs', () => {
+    const toggle = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'babysitMode')
+    const maxRounds = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'maxRounds')
+    const mentions = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'reviewMentions')
+
+    expect(toggle).toMatchObject({
+      type: 'switch',
+      defaultValue: false,
+      condition: { field: 'mode', value: 'cloud' },
+    })
+    expect(maxRounds).toMatchObject({
+      type: 'short-input',
+      defaultValue: '3',
+      mode: 'advanced',
+      condition: {
+        field: 'mode',
+        value: 'cloud',
+        and: { field: 'babysitMode', value: [true, 'true'] },
+      },
+    })
+    expect(mentions).toMatchObject({
+      type: 'short-input',
+      defaultValue: '',
+      hideDividerBefore: true,
+      required: {
+        field: 'mode',
+        value: 'cloud',
+        and: { field: 'babysitMode', value: [true, 'true'] },
+      },
+      condition: {
+        field: 'mode',
+        value: 'cloud',
+        and: { field: 'babysitMode', value: [true, 'true'] },
+      },
+    })
+    for (const output of [
+      'rounds',
+      'threadsClean',
+      'checksGreen',
+      'threadsResolved',
+      'commitsPushed',
+      'stopReason',
+    ]) {
+      expect(PiBlock.outputs[output]).toMatchObject({
+        condition: {
+          field: 'mode',
+          value: 'cloud',
+          and: { field: 'babysitMode', value: [true, 'true'] },
+        },
+      })
+    }
+  })
+
+  it('requires a task and hides Draft PR while Babysit Mode is enabled', () => {
+    const task = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'task')
+    const draft = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'draft')
+    const skills = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'skills')
+    const tools = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'tools')
+    const memory = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'memoryType')
+    const pullNumber = PiBlock.subBlocks.find((subBlock) => subBlock.id === 'pullNumber')
+
+    expect(task?.required).toBe(true)
+    expect(evaluateSubBlockCondition(draft?.condition, { mode: 'cloud' })).toBe(true)
+    expect(evaluateSubBlockCondition(draft?.condition, { mode: 'cloud', babysitMode: true })).toBe(
+      false
+    )
+    expect(
+      evaluateSubBlockCondition(draft?.condition, { mode: 'cloud', babysitMode: 'true' })
+    ).toBe(false)
+    expect(
+      evaluateSubBlockCondition(
+        PiBlock.subBlocks.find((subBlock) => subBlock.id === 'reviewMentions')?.condition,
+        { mode: 'cloud', babysitMode: 'true' }
+      )
+    ).toBe(true)
+    expect(evaluateSubBlockCondition(skills?.condition, { mode: 'cloud' })).toBe(true)
+    expect(evaluateSubBlockCondition(tools?.condition, { mode: 'cloud' })).toBe(false)
+    expect(evaluateSubBlockCondition(memory?.condition, { mode: 'cloud' })).toBe(true)
+    expect(evaluateSubBlockCondition(pullNumber?.condition, { mode: 'cloud' })).toBe(false)
+    expect(evaluateSubBlockCondition(pullNumber?.condition, { mode: 'cloud_review' })).toBe(true)
+  })
+})

apps/sim/blocks/blocks/pi.ts

@@ -19,6 +19,12 @@ interface PiResponse extends ToolResponse {
     branch?: string
     reviewUrl?: string
     commentsPosted?: number
+    rounds?: number
+    threadsClean?: boolean
+    checksGreen?: boolean
+    threadsResolved?: number
+    commitsPushed?: number
+    stopReason?: string
     tokens?: {
       input?: number
       output?: number
@@ -46,6 +52,31 @@ const CLOUD_ANY: { field: 'mode'; value: Array<'cloud' | 'cloud_review'> } = {
   field: 'mode',
   value: ['cloud', 'cloud_review'],
 }
+const BABYSIT_ENABLED_VALUES: Array<true | 'true'> = [true, 'true']
+const CLOUD_WITH_BABYSIT: {
+  field: 'mode'
+  value: 'cloud'
+  and: { field: 'babysitMode'; value: Array<true | 'true'> }
+} = {
+  field: 'mode',
+  value: 'cloud',
+  and: { field: 'babysitMode', value: BABYSIT_ENABLED_VALUES },
+}
+function getCloudWithoutBabysitCondition(values?: Record<string, unknown>): {
+  field: 'mode'
+  value: 'cloud'
+  and: { field: 'babysitMode'; value: true | 'true'; not: true }
+} {
+  return {
+    field: 'mode',
+    value: 'cloud',
+    and: {
+      field: 'babysitMode',
+      value: values?.babysitMode === 'true' ? 'true' : true,
+      not: true,
+    },
+  }
+}
 const LOCAL: { field: 'mode'; value: 'local' } = { field: 'mode', value: 'local' }
 const AUTHORING_MODES: { field: 'mode'; value: Array<'cloud' | 'local'> } = {
   field: 'mode',
@@ -85,9 +116,10 @@ export const PiBlock: BlockConfig<PiResponse> = {
   description: 'Run an autonomous coding agent on a repo',
   authMode: AuthMode.ApiKey,
   longDescription:
-    'The Pi Coding Agent runs the Pi harness against a real repository. Create PR spins up an isolated sandbox, clones a GitHub repo, edits with native shell + git, and opens a pull request. Review Code checks out a pinned PR snapshot with read-only tools and posts a structured review with optional inline comments. Local Dev edits files on your own machine over SSH. Create PR and Local Dev can reuse skills and multi-turn memory; Review Code runs without either because PR contents are untrusted. Any mode can optionally get one web_search tool backed by your own Exa, Serper, Parallel AI, or Firecrawl key; the agent writes its own queries, so repository content may reach the provider, and results are untrusted third-party data.',
+    'The Pi Coding Agent runs the Pi harness against a real repository. Create PR spins up an isolated sandbox, clones a GitHub repo, edits with native shell + git, and opens a pull request; Babysit Mode then keeps that pull request under watch, fixing trusted bot review threads and failing required checks in bounded rounds. Review Code checks out a pinned PR snapshot with read-only tools and posts a structured review with optional inline comments. Local Dev edits files on your own machine over SSH. Create PR and Local Dev can reuse skills and multi-turn memory; Review Code runs without either because PR contents are untrusted. Any mode can optionally get one web_search tool backed by your own Exa, Serper, Parallel AI, or Firecrawl key; the agent writes its own queries, so repository content may reach the provider, and results are untrusted third-party data.',
   bestPractices: `
   - Use Create PR for hands-off changes against a GitHub repo where a reviewable PR is the deliverable.
+  - Enable Babysit Mode on Create PR when trusted review bots and required checks should be monitored and fixed in bounded rounds.
   - Use Review Code to analyze an existing PR and leave summary + inline review comments.
   - Use Local Dev to edit a repo on your own machine; expose the machine on a public hostname/tunnel so Sim can reach it over SSH.
   - Create PR requires your own provider API key because the model runs in the sandbox. Review Code keeps the model key in Sim and can use either BYOK or a hosted key.
@@ -204,7 +236,7 @@ export const PiBlock: BlockConfig<PiResponse> = {
       paramVisibility: 'user-only',
       placeholder: 'GitHub personal access token',
       tooltip:
-        'Personal access token used for GitHub access. Create PR needs clone/push/PR permissions; Review Code needs clone + review permissions.',
+        'Personal access token used for GitHub access. Create PR needs clone/push/PR permissions; with Babysit Mode it also needs check/Actions reads, thread writes, and issue comments. Review Code needs clone + review permissions.',
       required: true,
       condition: CLOUD_ANY,
     },
@@ -216,6 +248,27 @@ export const PiBlock: BlockConfig<PiResponse> = {
       tooltip: 'The branch the pull request is opened against; the repo is cloned from it too.',
       condition: CLOUD,
     },
+    {
+      id: 'babysitMode',
+      title: 'Babysit Mode',
+      type: 'switch',
+      defaultValue: false,
+      description:
+        'Create the PR ready for review, request the configured bot reviews, and fix trusted feedback and required checks in bounded rounds.',
+      condition: CLOUD,
+    },
+    {
+      id: 'reviewMentions',
+      title: 'Reviewer Mentions',
+      type: 'short-input',
+      defaultValue: '',
+      placeholder: '@greptile, @cursor review',
+      tooltip:
+        'Required comma-separated issue comments. Each is posted after PR creation and again after every pushed Babysit fix.',
+      hideDividerBefore: true,
+      required: CLOUD_WITH_BABYSIT,
+      condition: CLOUD_WITH_BABYSIT,
+    },
     {
       id: 'branchName',
       title: 'Branch Name',
@@ -230,7 +283,7 @@ export const PiBlock: BlockConfig<PiResponse> = {
       type: 'switch',
       defaultValue: true,
       mode: 'advanced',
-      condition: CLOUD,
+      condition: getCloudWithoutBabysitCondition,
     },
     {
       id: 'prTitle',
@@ -269,6 +322,16 @@ export const PiBlock: BlockConfig<PiResponse> = {
         'How GitHub records the submitted review. Comment is neutral; Request changes marks the pull request as changes requested.',
       condition: CLOUD_REVIEW,
     },
+    {
+      id: 'maxRounds',
+      title: 'Maximum Rounds',
+      type: 'short-input',
+      defaultValue: '3',
+      placeholder: '3',
+      tooltip: 'Maximum number of agent fixing rounds, from 1 to 10.',
+      mode: 'advanced',
+      condition: CLOUD_WITH_BABYSIT,
+    },
 
     {
       id: 'host',
@@ -464,15 +527,28 @@ export const PiBlock: BlockConfig<PiResponse> = {
     },
     task: { type: 'string', description: 'Instruction for the coding agent' },
     model: { type: 'string', description: 'AI model to use' },
-    owner: { type: 'string', description: 'GitHub repository owner (Create PR and Review Code)' },
-    repo: { type: 'string', description: 'GitHub repository name (Create PR and Review Code)' },
-    githubToken: { type: 'string', description: 'GitHub token (Create PR and Review Code)' },
+    owner: { type: 'string', description: 'GitHub repository owner' },
+    repo: { type: 'string', description: 'GitHub repository name' },
+    githubToken: { type: 'string', description: 'GitHub token' },
     baseBranch: { type: 'string', description: 'Base branch for the PR (Create PR)' },
     branchName: { type: 'string', description: 'Branch to create (Create PR)' },
     draft: { type: 'boolean', description: 'Open the PR as a draft (Create PR)' },
     prTitle: { type: 'string', description: 'Pull request title (Create PR)' },
     prBody: { type: 'string', description: 'Pull request body (Create PR)' },
+    babysitMode: {
+      type: 'boolean',
+      description: 'Create the PR and babysit trusted bot reviews and required checks',
+    },
     pullNumber: { type: 'number', description: 'Pull request number (Review Code)' },
+    maxRounds: {
+      type: 'number',
+      description: 'Maximum Create PR Babysit fixing rounds (1-10)',
+    },
+    reviewMentions: {
+      type: 'string',
+      description:
+        'Required comma-separated bot review comments posted initially and after Babysit pushes',
+    },
     reviewEvent: {
       type: 'string',
       description: 'GitHub review event: COMMENT or REQUEST_CHANGES',
@@ -524,6 +600,36 @@ export const PiBlock: BlockConfig<PiResponse> = {
       description: 'Number of inline review comments posted',
       condition: CLOUD_REVIEW,
     },
+    rounds: {
+      type: 'number',
+      description: 'Babysit fixing rounds consumed',
+      condition: CLOUD_WITH_BABYSIT,
+    },
+    threadsClean: {
+      type: 'boolean',
+      description: 'Whether all actionable review threads are resolved',
+      condition: CLOUD_WITH_BABYSIT,
+    },
+    checksGreen: {
+      type: 'boolean',
+      description: 'Whether required checks are green with none pending',
+      condition: CLOUD_WITH_BABYSIT,
+    },
+    threadsResolved: {
+      type: 'number',
+      description: 'Review threads resolved by Babysit',
+      condition: CLOUD_WITH_BABYSIT,
+    },
+    commitsPushed: {
+      type: 'number',
+      description: 'Commits pushed by Babysit',
+      condition: CLOUD_WITH_BABYSIT,
+    },
+    stopReason: {
+      type: 'string',
+      description: 'Why the Babysit run stopped',
+      condition: CLOUD_WITH_BABYSIT,
+    },
     tokens: { type: 'json', description: 'Token usage statistics' },
     cost: { type: 'json', description: 'Cost of the run' },
     providerTiming: { type: 'json', description: 'Provider timing information' },

apps/sim/blocks/types.ts

@@ -314,6 +314,7 @@ export interface SubBlockConfig {
   connectionDroppable?: boolean
   hidden?: boolean
   hideFromPreview?: boolean // Hide this subblock from the workflow block preview
+  hideDividerBefore?: boolean // Visually group this field with the preceding visible subblock
   showWhenEnvSet?: string // Show this subblock only when the named NEXT_PUBLIC_ env var is truthy
   hideWhenHosted?: boolean // Hide this subblock when running on hosted sim
   hideWhenEnvSet?: string // Hide this subblock when the named NEXT_PUBLIC_ env var is truthy