The Agent Skills Directory

[PROMPT_INJECTION]: No evidence of instructions designed to bypass safety filters or override agent behavior was detected. The orchestration logic is task-oriented and well-structured.
[CREDENTIALS_UNSAFE]: The skill implements proactive measures to prevent sensitive data leakage. It explicitly warns users against pasting private keys into the chat and directs them to use secure alternatives like environment variables or encrypted Open Wallet Standard (OWS) vaults. It also recommends writing machine keys to local files to avoid exposure in command-line arguments.
[EXTERNAL_DOWNLOADS]: All external dependencies and resources, such as the peaq-os-cli and contract address configurations, are sourced from official and trusted repositories belonging to the vendor (peaqnetwork) or well-known infrastructure providers.
[COMMAND_EXECUTION]: Shell command execution is limited to environment checks and managing the machine onboarding process through the peaqos CLI. The skill provides clear narration and verification steps for each command, ensuring user awareness of all automated actions.

peaqos