The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructs the agent to execute echo $SCALEKIT_CLIENT_SECRET. This command prints production secrets directly to the standard output, allowing sensitive credentials to be captured in system logs, terminal history, or the agent's interaction records.\n- [COMMAND_EXECUTION]: The skill utilizes shell commands such as bash, curl, and rg to verify environment variables, perform network tests against authentication endpoints, and scan the local file system for hardcoded secrets.\n- [EXTERNAL_DOWNLOADS]: The skill performs network requests to verify configurations against scalekit.com and references external documentation from docs.scalekit.com, mcp.scalekit.com, and fonts.googleapis.com. These operations target the vendor's official infrastructure and well-known services.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted local data by running rg (ripgrep) to search project files. The capability inventory includes network access (curl) and command execution (bash), creating a surface where malicious content within the searched files could potentially influence agent behavior if the output is not strictly validated.

check-saaskit-prod