feat(agent/telemetry): Distinguish between production and dev environment based on VCS state

- Added a helper function `.app.utils.vcs_state_diverges_from_master()`. This function determines whether the relevant part of the codebase diverges from our `master`.
- Updated `.app.telemetry._setup_sentry()` to determine the default environment name using `vcs_state_diverges_from_master`.

autogpts/autogpt/autogpt/app/telemetry.py

@@ -3,7 +3,12 @@ import os
 import click
 from colorama import Fore, Style
 
-from .utils import env_file_exists, get_git_user_email, set_env_config_value
+from .utils import (
+    env_file_exists,
+    get_git_user_email,
+    set_env_config_value,
+    vcs_state_diverges_from_master,
+)
 
 
 def setup_telemetry() -> None:
@@ -49,7 +54,10 @@ def _setup_sentry() -> None:
     sentry_sdk.init(
         dsn="https://dc266f2f7a2381194d1c0fa36dff67d8@o4505260022104064.ingest.sentry.io/4506739844710400",  # noqa
         enable_tracing=True,
-        environment=os.getenv("TELEMETRY_ENVIRONMENT"),
+        environment=os.getenv(
+            "TELEMETRY_ENVIRONMENT",
+            "production" if not vcs_state_diverges_from_master() else "dev",
+        ),
     )
 
     # Allow Sentry to distinguish between users

autogpts/autogpt/autogpt/app/utils.py

@@ -1,3 +1,4 @@
+import contextlib
 import logging
 import os
 import re
@@ -84,6 +85,50 @@ def get_current_git_branch() -> str:
         return ""
 
 
+def vcs_state_diverges_from_master() -> bool:
+    """
+    Returns whether a git repo is present and contains changes that are not in `master`.
+    """
+    paths_we_care_about = "autogpts/autogpt/autogpt/**/*.py"
+    try:
+        repo = Repo(search_parent_directories=True)
+
+        # Check for uncommitted changes in the specified path
+        uncommitted_changes = repo.index.diff(None, paths=paths_we_care_about)
+        if uncommitted_changes:
+            return True
+
+        # Find OG AutoGPT remote
+        for remote in repo.remotes:
+            if remote.url.endswith(
+                tuple(
+                    # All permutations of old/new repo name and HTTP(S)/Git URLs
+                    f"{prefix}{path}"
+                    for prefix in ("://github.com/", "git@github.com:")
+                    for path in (
+                        f"Significant-Gravitas/{n}.git" for n in ("AutoGPT", "Auto-GPT")
+                    )
+                )
+            ):
+                og_remote = remote
+                break
+        else:
+            # Original AutoGPT remote is not configured: assume local codebase diverges
+            return True
+
+        master_branch = og_remote.refs.master
+        with contextlib.suppress(StopIteration):
+            next(repo.iter_commits(f"HEAD..{master_branch}", paths=paths_we_care_about))
+            # Local repo is one or more commits ahead of OG AutoGPT master branch
+            return True
+
+        # Relevant part of the codebase is on master
+        return False
+    except InvalidGitRepositoryError:
+        # No git repo present: assume codebase is a clean download
+        return False
+
+
 def get_git_user_email() -> str:
     try:
         repo = Repo(search_parent_directories=True)