The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill provides instructions to access and verify sensitive credentials, such as executing docker exec caddy env | grep CLOUDFLARE_API_KEY to check configuration. It also involves copying the /home/dawiddutoit/projects/network/.env file, which is documented to contain sensitive API tokens. These actions are within the scope of administrative maintenance and are accompanied by warnings against insecure storage.
[DATA_EXFILTRATION]: The skill performs local backup operations on sensitive file paths, including SSL private keys stored in the network_caddy_data Docker volume (/var/lib/docker/volumes/network_caddy_data/_data) and the project's .env file. These files are moved to a local backup directory (/home/dawiddutoit/projects/network/backups) as part of the skill's intended purpose.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data by reading and processing Docker container logs (docker logs caddy) to monitor renewal status, which represents a potential attack surface if malicious content is logged.
Ingestion points: Reading Caddy container logs via the docker logs command (SKILL.md).
Boundary markers: None identified.
Capability inventory: Bash command execution, filesystem read/write, and Docker CLI management.
Sanitization: The skill uses grep for pattern filtering, but does not perform content sanitization of the log data before processing.

caddy-certificate-maintenance