The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions in references/BUNDLED_SCRIPTS.md and assets/enhanced-quick-start.md to download and execute remote shell scripts directly using curl | bash from a GitHub repository containing a placeholder organization (YOUR-ORG).
[PRIVILEGE_ESCALATION]: In references/INSTALLATION.md, the skill performs high-severity system modifications, including replacing the root user's authorized_keys file and modifying /etc/ssh/sshd_config to enable PermitRootLogin with prohibit-password.
[INDIRECT_PROMPT_INJECTION]: The skill has a significant attack surface for indirect prompt injection.
Ingestion points: Data is ingested via a mandatory checklist in SKILL.md and environment templates (assets/env-enhanced.example).
Boundary markers: There are no boundary markers or instructions to ignore embedded commands in the provided data.
Capability inventory: The skill uses ssh, curl, sudo, docker exec, and systemctl across multiple scripts.
Sanitization: There is no evidence of validation or sanitization for parameters like COOLIFY_SERVER_IP, SSH_USER, or COOLIFY_INSTANCE_DOMAIN before they are interpolated into shell commands.
[COMMAND_EXECUTION]: The skill relies extensively on executing complex shell commands on remote servers via SSH, which is a sensitive capability that could be abused if parameters are manipulated.
[EXTERNAL_DOWNLOADS]: The skill fetches several external resources:
Downloads the official Coolify installer from cdn.coollabs.io.
Fetches the cloudflared binary from Cloudflare's official GitHub releases.
Downloads Docker installation components from download.docker.com.

coolify