The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install the official Alchemy CLI package (@alchemy/cli) from the public npm registry. This is a standard installation for a vendor-provided tool.
[COMMAND_EXECUTION]: The skill relies on executing shell commands through the alchemy CLI to perform blockchain operations and administrative tasks. It provides specific execution rules, such as using the --json and --no-interactive flags, to ensure predictable and parseable results.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill includes instructions for managing Alchemy API keys. It follows security best practices by recommending that these keys be stored in a local .env file and explicitly added to .gitignore. Furthermore, it contains a strict instruction to never display the raw API key values in the conversation output, treating them as sensitive credentials.
[INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data from the blockchain (such as NFT metadata and transaction logs). While this presents a theoretical attack surface where malicious data could influence agent behavior, the risk is minimized by the use of structured JSON output and the specific instructions provided for task-based command execution.

alchemy-cli