Skip to main content

​
sentry issue plan

Generate a solution plan for a Sentry issue using Seer AI. This command automatically runs root cause analysis if needed, then generates a solution plan with specific implementation steps to fix the issue.

​
Usage

sentry issue plan <issue-id>

​
Issue ID Formats

The plan command accepts the same issue identifier formats as explain and view: 
# Explicit org + ID
sentry issue plan sentry/EXTENSION-7 --cause 0

# Project + suffix
sentry issue plan cli-G --cause 1

# Short ID (searches across orgs)
sentry issue plan CLI-G --cause 0

# Suffix only (requires DSN context) 
sentry issue plan G --cause 0

# Numeric ID
sentry issue plan 123456789 --cause 0

​
Parameters

​
issue-id
string
required
Issue identifier in any supported format.

​
Flags

​
--cause
number
Root cause ID to create a plan for. Required if multiple root causes were identified by sentry issue explain.The cause ID corresponds to the numbered root causes from the explain output (0, 1, 2, …).Example:
sentry issue plan 123456789 --cause 0
sentry issue plan EXTENSION-7 --cause 1
​
--force
boolean
default:"false"
Force a new plan even if one already exists. Useful for regenerating plans after code changes.Example:
sentry issue plan 123456789 --cause 0 --force
​
--json
boolean
default:"false"
Output results as JSON including run ID, status, and solution data.Example:
sentry issue plan EXTENSION-7 --cause 0 --json

​
Examples

​
Basic Usage

# Generate plan for single root cause (auto-selected)
sentry issue plan 123456789

# Generate plan for specific cause when multiple exist
sentry issue plan 123456789 --cause 0

# With explicit org
sentry issue plan sentry/EXTENSION-7 --cause 1

​
Force Regeneration

# Regenerate plan with fresh analysis
sentry issue plan 123456789 --cause 0 --force

​
JSON Output

# Export to JSON
sentry issue plan EXTENSION-7 --cause 0 --json > solution.json

# Extract implementation steps
sentry issue plan CLI-G --cause 0 --json | jq '.solution.steps'

​
Sample Output

​
Human-Readable Format

Creating plan for cause #0...
"Undefined variable access in processData"

Generating solution... (this may take a few minutes)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Solution Plan
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Overview:
  Add defensive null checking before calling .map() on the users array.
  Ensure the API response is validated before processing.

Implementation Steps:

  1. Add null check in processData function
     File: src/components/UserList.js
     
     Replace:
       const userElements = users.map(user => renderUser(user));
     
     With:
       const userElements = (users || []).map(user => renderUser(user));
  
  2. Add API response validation
     File: src/api/client.js
     
     Add validation after API call:
       if (!response.data || !Array.isArray(response.data.users)) {
         throw new Error('Invalid API response: missing users array');
       }
  
  3. Add error boundary around UserList component
     File: src/pages/Dashboard.js
     
     Wrap component with error handling:
       <ErrorBoundary fallback={<UserListError />}>
         <UserList />
       </ErrorBoundary>

Testing:
  - Test with valid API response
  - Test with API response missing users field
  - Test with null/undefined users value
  - Verify error boundary displays fallback UI

PR: https://github.com/your-org/your-repo/pull/123

​
JSON Format

{
  "run_id": "autofix_abc123def456",
  "status": "COMPLETED",
  "solution": {
    "overview": "Add defensive null checking before calling .map() on the users array. Ensure the API response is validated before processing.",
    "steps": [
      {
        "id": 1,
        "title": "Add null check in processData function",
        "file": "src/components/UserList.js",
        "diff": "- const userElements = users.map(user => renderUser(user));\n+ const userElements = (users || []).map(user => renderUser(user));"
      },
      {
        "id": 2,
        "title": "Add API response validation",
        "file": "src/api/client.js",
        "code": "if (!response.data || !Array.isArray(response.data.users)) {\n  throw new Error('Invalid API response: missing users array');\n}"
      },
      {
        "id": 3,
        "title": "Add error boundary around UserList component",
        "file": "src/pages/Dashboard.js",
        "code": "<ErrorBoundary fallback={<UserListError />}>\n  <UserList />\n</ErrorBoundary>"
      }
    ],
    "testing": [
      "Test with valid API response",
      "Test with API response missing users field",
      "Test with null/undefined users value",
      "Verify error boundary displays fallback UI"
    ],
    "pr_url": "https://github.com/your-org/your-repo/pull/123"
  }
}

​
How It Works

  1. Root Cause Analysis: If not already done, runs sentry issue explain first
  2. Validate Causes: Ensures root causes were identified and validates cause selection
  3. Trigger Planning: Sends the selected root cause to Seer AI for solution planning
  4. Poll Progress: Monitors the planning process with status updates
  5. Extract Solution: Retrieves and formats the generated solution plan
The planning process may take several minutes as Seer analyzes code context, generates fixes, and validates the solution.

​
Requirements

GitHub integration and code mappings required. Seer AI needs access to your source code to generate accurate solution plans.

​
Prerequisites

  • Seer AI enabled in organization settings
  • GitHub integration configured for your organization
  • Code mappings set up for your project
  • Available Seer AI budget/quota

​
Setup Steps

  1. Enable Seer AI: Go to Organization Settings → Processing → AI Tools
  2. Configure GitHub: Organization Settings → Integrations → GitHub
  3. Set up code mappings: Project Settings → Integrations → GitHub → Code Mappings

​
Multiple Root Causes

If sentry issue explain identifies multiple root causes, you must specify which one to plan for: 
# Error if multiple causes exist and --cause not specified:
Multiple root causes found. Please specify one with --cause <id>:

  0: Undefined variable access in processData function
  1: Missing error handling in API client
  2: Incorrect response validation logic

Example: sentry issue plan 123456789 --cause 0
Select the cause ID that best matches the issue you want to fix first.

​
Notes

  • If only one root cause exists, --cause is optional and auto-selected.
  • The --force flag triggers a fresh plan generation, useful after code changes or to try alternative solutions.
  • Plans are cached per issue+cause combination to avoid redundant API calls.
  • The solution may include a PR link if Seer AI created a pull request with the proposed changes.
  • Organization context is automatically resolved from the issue ID.

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us