The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill manages sensitive blockchain credentials and provides a wallet-export-key command that displays the raw private key to the user (SKILL.md).\n- [EXTERNAL_DOWNLOADS]: The skill instructions trigger a download and execution of the Teneo CLI tool from the public npm registry using npx -y @teneo-protocol/cli if the local binary is not present.\n- [PERSISTENCE]: The agent deploy workflow installs new agents as background services that persist across reboots, utilizing system service managers such as systemd on Linux and launchd on macOS.\n- [COMMAND_EXECUTION]: The skill makes extensive use of shell command execution for its primary workflows, including managing a background daemon and interacting with the local filesystem.\n- [DYNAMIC_EXECUTION]: The agent deploy command performs runtime compilation of Go source code into an executable binary before service installation.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from various external network agents which may contain untrusted information.\n
Ingestion points: External data is ingested from third-party agents via the teneo command interface (SKILL.md).\n
Boundary markers: Missing. The skill instructions do not require the use of delimiters or 'ignore' instructions when returning agent output to the user context.\n
Capability inventory: The skill possesses capabilities for shell command execution, system service management, WebSocket network communication, and blockchain transaction signing (SKILL.md).\n
Sanitization: Missing. There are no defined procedures in the skill instructions for sanitizing or escaping the content returned by external agents before it is processed by the AI.

teneo-cli