The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute powerful system commands including tar, gpg, and sha256sum with sudo privileges to manage system-level archives in directories like /etc and /var/www.- [CREDENTIALS_UNSAFE]: The documentation references the sensitive file path ~/.backup-encryption-key for storing GPG symmetric passphrases. While it provides guidance on key hygiene, this path represents a potential target for credential access.- [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting data from the host filesystem into the agent's context during archive creation and verification.
Ingestion points: Files and directory trees from the local filesystem (e.g., /var/www, /etc) are read by the tar command.
Boundary markers: The instructions do not specify explicit boundary markers or sanitization for the content of the files being processed.
Capability inventory: The skill utilizes tar for filesystem reads/writes, sha256sum for integrity checks, and gpg for encryption/signing, all of which may be run with sudo elevation.
Sanitization: There is no evidence of sanitization or filtering of the file content before it is processed or reported by the agent.

linux-archive-integrity