Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward Browser Use API wrapper, but users should remember that browsing tasks and results go through a third-party cloud service.
Install only if you are comfortable sending browser automation instructions, visited-page data, extracted results, and screenshots to Browser Use's cloud service. Avoid using it for credentials, private/internal sites, regulated data, or irreversible form submissions unless you have separately approved that workflow and understand the provider's retention and billing behavior.
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
fi
# Submit task
RESPONSE=$(curl -s -X POST "$API_URL/tasks" \
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY" \
-H "Content-Type: application/json" \
-d "{\"task\": $(echo "$TASK" | python3 -c 'import json,sys; print(json.dumps(sys.stdin.read().strip()))')}")
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.
# Poll for completion
ELAPSED=0
while [[ $ELAPSED -lt $MAX_WAIT ]]; do
RESULT=$(curl -s "$API_URL/tasks/$TASK_ID" \
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY")
STATUS=$(echo "$RESULT" | python3 -c "import sys,json; print(json.load(sys.stdin).get('status',''))" 2>/dev/null || echo "")
The skill documents shell-based execution paths (curl commands and a helper script) but does not declare an explicit tool scope such as permissions or allowed-tools. That increases the chance the host agent may invoke broader shell capabilities than intended, which is risky for a skill that also transmits user-supplied tasks to an external service.
The trigger phrases include broad terms like scrape website, automate web task, and situations where local browser isn't available, which can cause unintended activation for many unrelated browsing requests. In this skill's context, accidental activation is more dangerous because it routes tasks to a third-party cloud browser service and may expose user instructions or page data externally.
The skill describes cloud browser automation but does not prominently warn that user tasks, visited pages, screenshots, and possibly extracted data are sent to and processed by an external cloud service. Because the API responses explicitly mention steps with screenshots, the privacy and data-handling implications are material and should be disclosed before use.
This command sends a natural-language task to api.browser-use.com, which is an intentional external transmission of user-provided instructions and potentially sensitive browsing objectives. In this skill's context, the transmission is core functionality, but it is still security-relevant because the remote service can receive confidential prompts or be directed toward sensitive sites.
# Submit task
curl -s -X POST https://api.browser-use.com/api/v2/tasks \
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"task": "Go to example.com and extract the main heading"}'
This command sends a natural-language task to api.browser-use.com, which is an intentional external transmission of user-provided instructions and potentially sensitive browsing objectives. In this skill's context, the transmission is core functionality, but it is still security-relevant because the remote service can receive confidential prompts or be directed toward sensitive sites.
# Submit task
curl -s -X POST https://api.browser-use.com/api/v2/tasks \
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"task": "Go to example.com and extract the main heading"}'
Polling task results from the remote service retrieves outputs and metadata from a cloud execution environment, confirming that browsing activity and resulting data reside off-platform. This is relevant because results may include sensitive extracted content and execution traces from prior external browsing steps.
-d '{"task": "Go to example.com and extract the main heading"}'
curl -s "https://api.browser-use.com/api/v2/tasks/TASK_ID"
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
set -euo pipefail
API_URL="https://api.browser-use.com/api/v2"
POLL_INTERVAL=5
MAX_WAIT=300
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
set -euo pipefail
API_URL="https://api.browser-use.com/api/v2"
POLL_INTERVAL=5
MAX_WAIT=300
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
set -euo pipefail
API_URL="https://api.browser-use.com/api/v2"
POLL_INTERVAL=5
MAX_WAIT=300
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
set -euo pipefail
API_URL="https://api.browser-use.com/api/v2"
POLL_INTERVAL=5
MAX_WAIT=300
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
set -euo pipefail
API_URL="https://api.browser-use.com/api/v2"
POLL_INTERVAL=5
MAX_WAIT=300
This POST request transmits user-supplied task data and an API credential to an external service. While this is core to the skill's purpose, it is still a genuine data-exposure boundary: any sensitive information included in the task leaves the local environment and is processed by a third party.
fi
# Submit task
RESPONSE=$(curl -s -X POST "$API_URL/tasks" \
-H "X-Browser-Use-API-Key: $BROWSER_USE_API_KEY" \
-H "Content-Type: application/json" \
-d "{\"task\": $(echo "$TASK" | python3 -c 'import json,sys; print(json.dumps(sys.stdin.read().strip()))')}")
The script sends the user-provided task string directly to a third-party cloud API, but it gives no explicit warning, consent prompt, or data-classification guidance before transmission. In a browser automation skill, task text may contain URLs, credentials, session details, internal targets, or other sensitive instructions, so silent exfiltration to an external service is a real privacy and security risk.
No suspicious patterns detected.