Never print or output environment variables (POSTHOG_API_KEY, POSTHOG_PROJECT_ID, POSTHOG_HOST, GH_PAT) to the terminal or console. These are stored securely in the Continue Hub and should never be displayed in plain text.

- Always authenticate API requests with the required headers:
  - Authorization: "Bearer {{POSTHOG_API_KEY}}"
- Do not fabricate or invent data if an API call fails.
- Limit outputs to analysis, issue tracking, and recommendations — not raw secrets or unrelated info.
 - Authenticate GitHub CLI using the `{{GH_PAT}}` token from Continue Hub:
    ```bash
    echo "{{GH_PAT}}" | gh auth login --with-token
    gh auth status
    ```
  - Always resolve the repository slug (`owner/repo`) **before any GitHub call**:
    1. If running in GitHub Actions, use `$GITHUB_REPOSITORY`.
    2. Else, run:
       ```bash
       git remote get-url origin | sed -E 's#.*github.com[:/](.+)\.git#\1#'
       ```
    3. If neither works, STOP and return: `❌ Missing repository information.`
  - **Never rely on the current working directory (CWD).**
    - Always pass `--repo "$REPO_SLUG"` explicitly to every `gh` command.
    - Never create or modify local files in the repository to generate issues.
  - Before creating any new issue:
    - Search for duplicates (open + recently closed).
    - Treat as duplicate if title/description are highly similar OR same page/flow + primary symptom.
    - If duplicate exists, add a comment with PostHog evidence instead of creating a new issue.
  - Never expose secret values in output.
  - Limit outputs to analysis summaries, issue creation results, and evidence reporting.
prompt: |
  After analyzing PostHog session data, automatically create GitHub issues for the **top 3 identified UX problems**, unless duplicates exist.

# Prompt
   Perform a comprehensive UX analysis and issue tracking based on PostHog session recordings.

   ## Analysis Process
   1. Fetch PostHog session recordings using:
      - Endpoint: `{{POSTHOG_HOST}}/api/projects/{{POSTHOG_PROJECT_ID}}/session_recordings/?limit=20`
      - Required headers (with valid secrets).
   2. Analyze data with the following priority:
      - Console errors (highest priority).
      - Session duration vs. activity score ratios.
      - Bounce rates (0–1 second sessions).
      - Click/interaction patterns.
      - Page-specific retention issues.

   ## Issue Identification
   - Identify the **top 3 UX issues** based on frequency and impact.
   - Categorize issues by severity:
   - **Critical**: console errors
   - **High**: engagement-related problems
   - **Medium**: retention issues
   - Provide **evidence-based recommendations** for each identified issue.

After analysis, automatically create GitHub issues based on the data in the GitHub repository Issues for the top 3 identified problems. The environment variable is in the continue Hub as {{GH_PAT}}.
1. Select the top 3 UX problems based on severity and impact. 2. For each candidate:
  - Set ISSUE_TITLE: "[UX Issue] {Brief Description}"
  - Search for duplicates (open + recently closed):
    gh issue list --repo "$REPO_SLUG" --state open --search "\"{Brief Description}\" in:title" --limit 10 --json number,title,url
    gh issue list --repo "$REPO_SLUG" --state closed --search "\"{Brief Description}\" in:title" --limit 10 --json number,title,url
  - Near-duplicate heuristic: treat as duplicate if same or highly similar brief description (case-insensitive) OR same affected page/flow and same primary symptom.
  - If duplicate exists:
    - Compose EVIDENCE_COMMENT with PostHog evidence (session IDs, metrics, repro steps).
    - Post comment:
      gh issue comment <ISSUE_NUMBER> --repo "$REPO_SLUG" --body "$EVIDENCE_COMMENT"
    - Skip creation.
  - If no duplicate:
    - Build ISSUE_BODY using the template below.
    - Create issue:
      ISSUE_URL=$(gh issue create --repo "$REPO_SLUG" --title "$ISSUE_TITLE" --body "$ISSUE_BODY" --label "automated" --label "$PRIMARY_LABEL" --label "$SEVERITY_LABEL" --json url,number --jq '.url')

Issue Body Template ## Problem Description {Detailed issue description}
## Impact Assessment - Affected sessions: {percentage/count} - Severity: {Critical|High|Medium} - User journey impact: {description}
## Evidence from PostHog - Sessions: {IDs or links} - Key events/metrics: {specific, numeric} - Repro path: {steps/URLs}
## Recommended Actions - [ ] {Action item 1} - [ ] {Action item 2} - [ ] {Action item 3}
## Links - PostHog dashboard/query: {URL} - Related issues (if any): #{id}, #{id}
Output Format - "🔍 Analyzed X PostHog sessions" - "🎯 Identified 3 UX issues" - "📋 Created GitHub issues: #XXX, #XXX, #XXX" (or "Updated duplicates: #YYY, #ZZZ") - Brief list: {issue title} — #{number} (URL)