Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward TON wallet lookup guide that uses TonScan, with a privacy documentation gap but no hidden, destructive, persistent, or privilege-seeking behavior.
Installers should understand that checking a TON address will contact TonScan and can link that address lookup to the user's context. Use it for explicit TON wallet lookups, avoid pasting addresses you do not want queried through TonScan, and prefer adding a short privacy notice if publishing an updated version.
The trigger guidance explicitly says to invoke the skill for casual phrases like 'check this address for me,' which is broader than a narrowly scoped blockchain lookup intent. Overbroad activation can cause the agent to send wallet addresses to an external service when the user did not clearly consent to a third-party lookup, increasing privacy and data-handling risk.
The skill description encourages use of a public third-party API but does not warn that wallet addresses supplied by users will be transmitted off-platform. Even though blockchain addresses are often public, they can still be sensitive in context because they may reveal a user's holdings, activity, or associations when linked to their identity.
This endpoint definition establishes that user-supplied wallet addresses are transmitted to an external third-party service. In this skill's context that behavior is functional, but it is still a real data egress path and becomes risky if used without clear disclosure and consent.
GET https://api.tonscan.com/api/bt/getAddressInformation?address={ADDRESS}
The curl example demonstrates direct transmission of an address to TonScan, reinforcing that the skill exports user-provided data to a public external service. Although expected for blockchain lookups, this is still a privacy-relevant behavior that should be treated as a true security concern in agent environments.
curl -s "https://api.tonscan.com/api/bt/getAddressInformation?address=EQDtFpEwcFAEcRe5mLVh2N6C0x-_hJEM7W61_JLnSF74p4q2" \
| jq '.json.data.detail.balance'
This example sends the address variable to the TonScan API and processes the returned account details, confirming an external data-sharing path. If a user pastes a personally associated wallet address, the lookup can expose sensitive financial metadata to a third party without adequate notice.
ADDRESS="EQDtFpEwcFAEcRe5mLVh2N6C0x-_hJEM7W61_JLnSF74p4q2"
curl -s "https://api.tonscan.com/api/bt/getAddressInformation?address=${ADDRESS}" \
| jq '{
address: .json.data.detail.address,
balance_nanoton: .json.data.detail.balance,
The Python sample performs an outbound request with the address as a parameter, which is a concrete external transmission mechanism. In a skill system, sample code often mirrors operational behavior, so this indicates real privacy exposure rather than a hypothetical issue.
import requests
def get_ton_balance(address: str) -> dict:
url = "https://api.tonscan.com/api/bt/getAddressInformation"
resp = requests.get(url, params={"address": address})
resp.raise_for_status()
The JavaScript example similarly transmits the address to TonScan via fetch, confirming the same third-party data egress path in another implementation. The skill context makes this expected, but not harmless: addresses can be privacy-sensitive once linked to a user or conversation context.
async function getTonBalance(address) {
const url = new URL("https://api.tonscan.com/api/bt/getAddressInformation");
url.searchParams.set("address", address);
const res = await fetch(url);
This example again shows an address being sent to TonScan, demonstrating repeated encouragement of external lookups without paired privacy messaging. Repetition across the skill increases the likelihood of routine data egress becoming normalized and overlooked by users.
Checking for empty/uninitialized accounts:
curl -s "https://api.tonscan.com/api/bt/getAddressInformation?address=..." \
| jq 'if .json.data.detail.balance == "0" then "Empty wallet" else "Has funds" end'
No suspicious patterns detected.