Google Maps

Google Maps integration for OpenClaw with Routes API. Use for: (1) Distance/travel time calculations with traffic prediction, (2) Turn-by-turn directions, (3...

Security Scan

This skill appears to do what it claims — a Google Maps / Routes API wrapper that only needs a Google API key and calls Google endpoints; I found no unexplained credential requests, hidden endpoints, or disproportionate installation steps.

Purpose & Capability

Name/description, the declared env var (GOOGLE_API_KEY), the documented APIs (Routes, Places, Geocoding) and the included Python code all align. The functionality (distance, directions, matrix, geocoding, places) matches the code and required APIs; there are no unrelated credentials or tools requested.

Instruction Scope

SKILL.md instructs running the included Python script and to set GOOGLE_API_KEY (and optional GOOGLE_MAPS_LANG). The instructions reference running python3 against lib/map_helper.py (and once reference a longer path 'skills/google-maps/lib/map_helper.py') — a minor path inconsistency but not security-critical. I inspected the start of map_helper.py; it performs only API calls to Google endpoints and local processing. I saw no instructions to read arbitrary host files, other env vars, or to exfiltrate data to unknown endpoints.

Install Mechanism

Registry metadata lists this as instruction-only (no formal install spec), but SKILL.md metadata includes an install entry to 'pip install requests'. The code imports requests, so that is reasonable. This is low-risk (standard PyPI dependency) but you should ensure requests is available in your runtime or install it from your trusted package manager rather than running arbitrary install commands from an untrusted source.

Credentials

Only GOOGLE_API_KEY (primary) and an optional fallback GOOGLE_MAPS_API_KEY plus GOOGLE_MAPS_LANG are referenced. These are appropriate and expected for a Maps integration. No unrelated secrets (AWS, GitHub tokens, etc.) are requested.

Persistence & Privilege

The skill is not marked always:true and does not request elevated or persistent platform privileges. It uses allowed-tools: exec to run the Python helper (expected). It does not modify other skills or system-wide configs in the files I saw.

Assessment

This skill looks coherent and limited to Google Maps functionality, but before installing: (1) Only provide a Google Maps API key with the minimal enabled APIs (Routes, Places, Geocoding) and quota/billing limits you accept. (2) Install the requests package from your trusted environment (do not run arbitrary pip commands supplied by unknown parties). (3) Review or run the included lib/map_helper.py in a sandbox if you want to audit behavior fully (the file is straightforward and calls official Google endpoints). (4) Note the skill can execute the helper via exec — only enable the skill for agents you trust. If you need higher assurance, ask the author for a full file review or a signed release/source repository.

Google Maps integration powered by the Routes API.

Requirements

• GOOGLE_API_KEY environment variable
• Enable in Google Cloud Console: Routes API, Places API, Geocoding API

Configuration

Env Variable	Default	Description
GOOGLE_API_KEY	-	Required. Your Google Maps API key
GOOGLE_MAPS_API_KEY	-	Alternative to GOOGLE_API_KEY (fallback)
GOOGLE_MAPS_LANG	en	Response language (en, he, ja, etc.)

Set in OpenClaw config:

{
  "env": {
    "GOOGLE_API_KEY": "AIza...",
    "GOOGLE_MAPS_LANG": "en"
  }
}

Script Location

python3 skills/google-maps/lib/map_helper.py <action> [options]

---

Actions

distance - Calculate travel time

python3 lib/map_helper.py distance "origin" "destination" [options]

Options:

Option	Values	Description
--mode	driving, walking, bicycling, transit	Travel mode (default: driving)
--depart	now, +30m, +1h, 14:00, 2026-02-07 08:00	Departure time
--arrive	14:00	Arrival time (transit only)
--traffic	best_guess, pessimistic, optimistic	Traffic model
--avoid	tolls, highways, ferries	Comma-separated

Examples:

python3 lib/map_helper.py distance "New York" "Boston"
python3 lib/map_helper.py distance "Los Angeles" "San Francisco" --depart="+1h"
python3 lib/map_helper.py distance "Chicago" "Detroit" --depart="08:00" --traffic=pessimistic
python3 lib/map_helper.py distance "London" "Manchester" --mode=transit --arrive="09:00"
python3 lib/map_helper.py distance "Paris" "Lyon" --avoid=tolls,highways

Response:

{
  "distance": "215.2 mi",
  "distance_meters": 346300,
  "duration": "3 hrs 45 mins",
  "duration_seconds": 13500,
  "static_duration": "3 hrs 30 mins",
  "duration_in_traffic": "3 hrs 45 mins"
}

---

directions - Turn-by-turn route

python3 lib/map_helper.py directions "origin" "destination" [options]

Additional options (beyond distance):

Option	Description
--alternatives	Return multiple routes
--waypoints	Intermediate stops (pipe-separated)
--optimize	Optimize waypoint order (TSP)

Examples:

python3 lib/map_helper.py directions "New York" "Washington DC"
python3 lib/map_helper.py directions "San Francisco" "Los Angeles" --alternatives
python3 lib/map_helper.py directions "Miami" "Orlando" --waypoints="Fort Lauderdale|West Palm Beach" --optimize

Response includes: summary, labels, duration, static_duration, warnings, steps[], optimized_waypoint_order

---

matrix - Distance matrix

Calculate distances between multiple origins and destinations:

python3 lib/map_helper.py matrix "orig1|orig2" "dest1|dest2"

Example:

python3 lib/map_helper.py matrix "New York|Boston" "Philadelphia|Washington DC"

Response:

{
  "origins": ["New York", "Boston"],
  "destinations": ["Philadelphia", "Washington DC"],
  "results": [
    {"origin_index": 0, "destination_index": 0, "distance": "97 mi", "duration": "1 hr 45 mins"},
    {"origin_index": 0, "destination_index": 1, "distance": "225 mi", "duration": "4 hrs 10 mins"}
  ]
}

---

geocode - Address to coordinates

python3 lib/map_helper.py geocode "1600 Amphitheatre Parkway, Mountain View, CA"
python3 lib/map_helper.py geocode "10 Downing Street, London"

reverse - Coordinates to address

python3 lib/map_helper.py reverse 40.7128 -74.0060  # New York City
python3 lib/map_helper.py reverse 51.5074 -0.1278  # London

---

search - Find places

python3 lib/map_helper.py search "coffee near Times Square"
python3 lib/map_helper.py search "pharmacy in San Francisco" --open

details - Place information

python3 lib/map_helper.py details "<place_id>"

---

Traffic Models

Model	Use Case
best_guess	Default balanced estimate
pessimistic	Important meetings (worst-case)
optimistic	Best-case scenario

---

Regional Notes

Some features may not be available in all countries:

Feature	Availability
--fuel-efficient	US, EU, select countries
--shorter	Limited availability
--mode=two_wheeler	Asia, select countries

Check Google Maps coverage for details.

---

Multilingual Support

Works with addresses in any language:

# Hebrew
python3 lib/map_helper.py distance "תל אביב" "ירושלים"
python3 lib/map_helper.py geocode "דיזנגוף 50, תל אביב"

# Japanese
python3 lib/map_helper.py distance "東京" "大阪"

# Arabic
python3 lib/map_helper.py distance "دبي" "أبو ظبي"

Language configuration:

1. Set default via env: GOOGLE_MAPS_LANG=he (persists)
2. Override per-request: --lang=ja

# Set Hebrew as default in OpenClaw config
GOOGLE_MAPS_LANG=he

# Override for specific request
python3 lib/map_helper.py distance "Tokyo" "Osaka" --lang=ja

---

Help

python3 lib/map_helper.py help