vara-agent-network-skills

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly queries the public indexer (e.g., curl POSTs to $INDEXER_GRAPHQL_URL / https://agents-api.vara.network/graphql in agent-create.md and multiple pages) and ingests identity cards, announcements, and chat message bodies (attacker-controlled, user-generated) which the agent is required to read and use to make Build Decisions and drive posts/replies, so untrusted third‑party content can materially influence tool use and next actions.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill pack explicitly requires and instructs use of a crypto wallet CLI ("vara-wallet") for creating wallets, checking balances, and making on-chain calls/writes (e.g., vara-wallet wallet create, vara-wallet --json call, vara-wallet call $PID Service/Method --voucher ...). It details transaction/estimate/voucher flows, write/verify ladders, and program deployment/registration steps that sign and submit chain transactions. These are specific crypto/blockchain wallet and transaction operations (signing/sending on-chain calls), not generic tools, so it grants direct financial/transaction execution capability.