chore: add codespell pre-commit hook to catch typos

Merge #6406

Add a codespell hook to catch typos automatically.
Fix existing typos flagged by codespell.

PiperOrigin-RevId: 952416492

Author: anxkhn

.pre-commit-config.yaml

@@ -72,3 +72,11 @@ repos:
         exclude: (?i)SKILL\.md$
         additional_dependencies:
           - mdformat-gfm
+  - repo: https://github.com/codespell-project/codespell
+    rev: v2.4.2
+    hooks:
+      - id: codespell
+        # Configuration (skip globs, ignored words) lives in the
+        # [tool.codespell] table in pyproject.toml.
+        additional_dependencies:
+          - tomli

contributing/samples/integrations/data_agent/agent.py

@@ -36,7 +36,7 @@
 CREDENTIALS_TYPE = None
 
 if CREDENTIALS_TYPE == AuthCredentialTypes.OAUTH2:
-  # Initiaze the tools to do interactive OAuth
+  # Initialize the tools to do interactive OAuth
   # The environment variables OAUTH_CLIENT_ID and OAUTH_CLIENT_SECRET
   # must be set
   credentials_config = DataAgentCredentialsConfig(

contributing/samples/integrations/oauth_calendar_agent/agent.py

@@ -42,7 +42,7 @@
 SCOPES = ["https://www.googleapis.com/auth/calendar"]
 
 calendar_toolset = CalendarToolset(
-    # you can also replace below customized `list_calendar_events` with build-in
+    # you can also replace below customized `list_calendar_events` with built-in
     # google calendar tool by adding `calendar_events_list` in the filter list
     client_id=oauth_client_id,
     client_secret=oauth_client_secret,

contributing/samples/integrations/spanner_rag_agent/README.md

@@ -296,7 +296,7 @@ There are a few options to perform similarity search:
            "from the Spanner database."
        ),
        instruction="""
-       You are a helpful assistant that answers user questions to find the most relavant information from a Spanner database.
+       You are a helpful assistant that answers user questions to find the most relevant information from a Spanner database.
        1. Always use the `similarity_search` tool to find relevant information.
        2. If no relevant information is found, say you don't know.
        3. Present all the relevant information naturally and well formatted in your response.

contributing/samples/workflows/route/agent.py

@@ -57,7 +57,7 @@ def route_on_category(category: InputCategory):
 
 def handle_other():
   yield Event(
-      message="Sorry I can only anwer questions or comment on statements."
+      message="Sorry I can only answer questions or comment on statements."
   )
 
 

pyproject.toml

@@ -112,6 +112,7 @@ optional-dependencies.db = [
   "sqlalchemy-spanner>=1.14",
 ]
 optional-dependencies.dev = [
+  "codespell[toml]==2.4.2",
   "flit>=3.10",
   # Lint tools used by scripts/run_precommit_checks.sh (which runs them directly,
   # without the pre-commit framework). Formatters that rewrite files are pinned
@@ -300,6 +301,17 @@ line_length = 200
 single_line_exclusions = []
 known_third_party = [ "a2a", "google.adk" ]
 
+[tool.codespell]
+# Real words/identifiers that codespell misreads as typos:
+#   hel/serie/strin -> substrings in test fixtures; te -> local variable;
+#   rouge -> the ROUGE metric; unparseable -> valid spelling variant;
+#   re-use/re-used -> intentional hyphenation; lamda -> Google LaMDA project.
+ignore-words-list = "hel,serie,strin,te,rouge,unparseable,re-use,re-used,lamda"
+# CHANGELOG.md is generated from commit messages; lockfiles, notebooks, JSON
+# fixtures, bundled JS/source maps, and the vendored CLI browser bundle are
+# generated or data files, not prose we own.
+skip = "*CHANGELOG.md,*.lock,*.ipynb,*.json,*.js,*.map,*/cli/browser/*"
+
 [tool.mypy]
 mypy_path = [ "src" ]
 exclude = [ "contributing/samples/", "tests/" ]

src/google/adk/evaluation/rubric_based_tool_use_quality_v1.py

@@ -136,7 +136,7 @@ class RubricBasedToolUseV1Evaluator(RubricBasedEvaluator):
   """An Evaluator for rubric based assessment of the agent's usage of Tools.
 
   Example: Lets take an example of a Weather Agent that has access to two tools:
-  1: GeoCoding Tool: Coverts a city name, address or zip code into geographic
+  1: GeoCoding Tool: Converts a city name, address or zip code into geographic
   coordinates.
   2: GetWeather Tool: Gets weather for the next 10 days for the given geographic
   coordinates.

src/google/adk/runners.py

@@ -1631,7 +1631,7 @@ async def run_live(
     *   **Other Control Events:** Most control events are saved.
 
     **Events Saved to the Session:**
-    *   **Live Model Audio Events with File Data:** Both input and ouput audio
+    *   **Live Model Audio Events with File Data:** Both input and output audio
         data are aggregated into an audio file saved into artifacts. The
         reference to the file is saved as event in the `file_data` to session
         if RunConfig.save_live_model_audio_to_session is True.

src/google/adk/tools/application_integration_tool/clients/connections_client.py

@@ -248,7 +248,7 @@ def get_connector_base_spec() -> Dict[str, Any]:
                 "host": {
                     "type": "string",
                     "default": "",
-                    "description": "Host name incase of tls service directory",
+                    "description": "Host name in case of tls service directory",
                 },
                 "entity": {
                     "type": "string",

src/google/adk/tools/bigquery/skills/bigquery-graph/SKILL.md

@@ -168,10 +168,10 @@ commas `,`. When multiple comma-separated patterns are used:
     **equijoin** on that variable.
 
 ```sql
--- Equijoin example where 'interm' connects the two paths
+-- Equijoin example where 'interim' connects the two paths
 GRAPH <project>.<dataset>.<graph>
-MATCH (src:Account)-[t1:Transfers]->(interm:Account),
-      (interm)<-[:Owns]-(p:Person)
+MATCH (src:Account)-[t1:Transfers]->(interim:Account),
+      (interim)<-[:Owns]-(p:Person)
 RETURN src.id AS account_id, p.name AS owner_name
 ```
 
@@ -393,7 +393,7 @@ standard GoogleSQL that you **MUST** adhere to.
 In BigQuery Graph, unlike standard GoogleSQL, you **MUST** specify the graph
 name within the subquery block. If the outer query uses `GRAPH
 <project>.<dataset>.<graph>`, the internal subquery must also explicitly
-re-declare it.
+redeclare it.
 
 ```sql
 MATCH (n1)