The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides commands to download and execute scripts directly from an external domain (herdr.dev) via curl piped to sh (Bash) and irm piped to iex (PowerShell). Piped execution from unverified domains is a significant security risk as it allows for immediate execution of remote code without prior review.
[COMMAND_EXECUTION]: The skill relies on executing various shell commands (herdr, curl, irm) and integrations which could be manipulated if the environment or external sources are compromised. It includes instructions to install integrations and perform updates based on tool output.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources that could contain malicious instructions.
Ingestion points: The skill links to and reads documentation from github.com/ogulcancelik/herdr (found in references/upstream-docs.md).
Boundary markers: No explicit boundary markers or ignore instructions warnings are present in the documentation handling instructions.
Capability inventory: The skill has the ability to execute shell commands, install integrations, and update the tool, which could be abused if malicious instructions are present in the external docs.
Sanitization: No evidence of input sanitization or validation for the content retrieved from external documentation.

herdr-setup