launkr

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text from wallet users via CLI options (e.g., --uri <uri> / --name <name> / --symbol <symbol>) and passes it into on-chain transaction payloads without any selection-by-item step.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill performs runtime calls to the Launkr API (e.g. https://launkr.io/api/launch and https://launkr.io/api/protocol?network=...) and uses the API's returned clarityCode and pool-creation args to deploy and invoke on-chain contracts, meaning remote content directly controls executed code and critical runtime behavior.

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

• Direct money access detected (high risk: 1.00). The skill is explicitly designed to perform on-chain financial actions: it deploys tokens, creates liquidity pools, and executes swaps that move STX and SIP-010 tokens. It requires a wallet and exposes write operations that sign and broadcast transactions (launch, create-pool, swap-buy, swap-sell, set-fee-receiver, etc.). These are direct crypto/blockchain transaction capabilities (wallet-based swaps and contract calls), i.e. direct financial execution.