The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The installation instructions direct users to download a ZIP archive from openclawby.com or clone a repository from GitHub to obtain the skill's executable components.
[CREDENTIALS_UNSAFE]: The skill documentation explicitly instructs users to store sensitive authentication data, such as Binance/OKX API secrets and private keys for Ethereum and Solana wallets, in environment variables or persistent shell configuration files like ~/.bashrc and ~/.zshrc.
[COMMAND_EXECUTION]: The skill involves the execution of local Python scripts (e.g., binance_connector.py, polymarket_connector.py) to perform trade operations and system setup.
[REMOTE_CODE_EXECUTION]: A self-update mechanism is described that monitors a remote manifest and applies code updates from a remote repository, which allows for the execution of updated code without a full manual review.
[DATA_EXFILTRATION]: The skill performs network operations to api.openclawby.com using the CLAWBY_API_KEY stored in environment variables, creating a pathway where sensitive credentials could be transmitted externally.
[PROMPT_INJECTION]: The skill ingests data from untrusted external sources, including real-time social media searches (Twitter/X and Facebook), which creates an indirect prompt injection surface that could influence agent behavior based on external content.

openclawby-skills