Security audit
Security checks for vulnerabilities and agentic risk
This skill is a clearly disclosed Hail integration for sending email, SMS, and voice calls, with expected third-party data transmission and no hidden local execution.
Install only if you want your agent to use Hail as an external communications provider. Treat emails, phone numbers, message bodies, call prompts, owner email, claim URLs, and the HAIL_API_KEY as sensitive. Require clear user approval and recipient consent before signup or any outbound email, SMS, or voice call, and avoid bulk or sensitive communications unless your legal/compliance basis is settled.
The skill explicitly enables sending email, SMS, and voice calls through a third-party service, but the description does not clearly warn users that recipient contact details, message content, and owner email will be transmitted to an external provider. This omission can lead to uninformed handling of sensitive personal data and unexpected privacy/compliance exposure.
The signup flow transmits the human owner's email address to an external service and instructs the agent to accept terms on the owner's behalf. Even though the text advises confirming approval first, this creates privacy and authorization risk because an agent could disclose personal data or bind a user to third-party terms without robust consent handling.
only set tou_accepted after confirming your owner is okay with that.
curl -s -X POST https://hail.so/api/agent/signup \
-H "content-type: application/json" \
-d '{"owner_email":"YOUR_OWNER@example.com","agent_name":"YOUR_NAME","tou_accepted":true,"ref":"clawhub"}'
This endpoint sends recipient email addresses and message content to an external communications provider. In context, that is the intended function of the skill, but it is still security-relevant because it can expose personal or sensitive content to a third party and can be abused for unauthorized outbound messaging if user consent and recipient validation are weak.
# Email
curl -s -X POST https://api.hail.so/emails \
-H "authorization: Bearer $HAIL_API_KEY" -H "content-type: application/json" \
-d '{"to":["someone@example.com"],"subject":"Hello from my agent","body_text":"Sent via Hail.","recipient_consent":true}'
This endpoint transmits phone numbers and SMS content to an external provider, which materially increases privacy and regulatory risk because SMS often involves personal contact data and consent requirements. In the hands of an autonomous agent, this can enable spam, harassment, or accidental disclosure if sends are not tightly governed.
-d '{"to":["someone@example.com"],"subject":"Hello from my agent","body_text":"Sent via Hail.","recipient_consent":true}'
# SMS
curl -s -X POST https://api.hail.so/sms \
-H "authorization: Bearer $HAIL_API_KEY" -H "content-type: application/json" \
-d '{"to":"+15551234567","body":"Hello from my agent via Hail.","recipient_consent":true}'
The voice-call API sends phone numbers and a system prompt to an external service that will generate or deliver a call on the agent's behalf. This is more dangerous in context because autonomous calling can facilitate harassment, impersonation, social engineering, or disclosure of sensitive content to unintended recipients.
-d '{"to":"+15551234567","body":"Hello from my agent via Hail.","recipient_consent":true}'
curl -s -X POST https://api.hail.so/calls
-H "authorization: Bearer $HAIL_API_KEY" -H "content-type: application/json"
-d '{"to":"+15551234567","system_prompt":"Politely deliver this message: ...","recipient_consent":true}'