The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides installation instructions for the Foundry toolkit that involve downloading a script from a well-known service domain (foundry.paradigm.xyz) and piping it directly to a shell for execution.
[COMMAND_EXECUTION]: The skill performs blockchain interactions by building and executing cast commands in the terminal. These commands are dynamically generated using parameters such as RPC URLs, token contract addresses, and transaction amounts.
[CREDENTIALS_UNSAFE]: Documentation within the skill guides users on setting up wallet access, including the use of environment variables and the creation of local password files (~/.foundry/.password). The skill author includes prominent warnings stating that these methods are intended for development purposes and pose security risks like plaintext exposure.
[INDIRECT_PROMPT_INJECTION]: The skill processes information from the KyberSwap Token API and CoinGecko to determine token contract addresses and decimals, which are subsequently used in the construction of executable commands.
Ingestion points: External data is retrieved from token-api.kyberswap.com, CoinGecko, and user-provided RPC endpoints.
Boundary markers: The skill requires manual user confirmation ('yes/no') before executing approval transactions, which serves as a procedural boundary.
Capability inventory: The skill possesses the capability to execute shell commands for both reading state (cast call) and sending transactions (cast send).
Sanitization: The skill validates the sender's wallet address using a regular expression (^0x[a-fA-F0-9]{40}$), but there is no explicit sanitization for the token addresses or metadata returned from external APIs before they are interpolated into commands.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to external APIs for token resolution and provides links to download third-party software necessary for its operation.

swap-approve