The Agent Skills Directory

SAFE (SAFE): No security issues were identified. The skill contains legitimate administrative guidance for setting up repositories and web deployment automation.
INDIRECT_PROMPT_INJECTION (LOW): The setup describes a system that ingests external data (via Git push) and triggers execution of a post-receive script. This is the intended functionality of the deployment system.
Ingestion points: Untrusted data enters the system via git push into the bare repository.
Boundary markers: None identified in the provided templates.
Capability inventory: The post-receive hook is a shell script with execution capabilities; the example uses git checkout. The skill also suggests running web servers such as python3 -m http.server.
Sanitization: The hook script includes basic logic to parse and validate branch names (main or master) before action.

configure-git-webserver