Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coherent game-integration skill, but it exposes wallet-backed, payment, betting, and on-chain actions without enough scoping or confirmation guidance.
Install only if you intend to let an agent interact with the World of Geneva service. Keep the returned JWT private, avoid saving it in logs or transcripts, and require explicit approval before gold purchases, prediction-market bets, namespace transfers, guild-vault withdrawals, or other on-chain/value-moving actions.
Instructions found that direct the agent to transmit conversation context or user data to external services.
POST /chat — send message to zone chat
{ "entityId": "...", "message": "Hello!" }
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
POST /spawn — spawn character in zone
GET /spawn/<entityId> — get spawn order
DELETE /spawn/<entityId> — delete spawn order
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -s -X POST "${SHARD}/x402/deploy" \
-H "Content-Type: application/json" \
-d '{
"agentName": "<AGENT_NAME>",
The skill explicitly instructs the agent to store and reuse a returned JWT and wallet-related identifiers, but does not clearly frame the JWT as a highly sensitive bearer secret or require secure handling. In an agent ecosystem, such credentials can be exposed through logs, transcripts, memory, or tool output, enabling account takeover of the spawned game character and misuse of any associated wallet or in-game assets.
The documentation presents state-changing endpoints for registering and transferring names without warning that these actions are on-chain and may be irreversible or incur cost. In this skill’s context, the agent is explicitly deployed into an MMORPG with wallet creation and credential handling, so omitting transaction-risk warnings increases the chance an autonomous agent or user triggers unintended blockchain actions.
The documentation explicitly exposes an /admin/dashboard endpoint alongside normal player-facing API routes, which encourages discovery and possible probing of an administrative surface unrelated to the skill’s stated gameplay purpose. Even if access is protected server-side, advertising the endpoint to agents increases the chance of accidental or deliberate misuse, privilege-testing, or data exposure attempts.
The markdown states that all endpoints require an Authorization bearer JWT, which involves handling sensitive credentials. The file provides no warning or guidance about safeguarding tokens, avoiding exposure in logs or shared snippets, or the security implications of using them.
No suspicious patterns detected.