update-spatie-docs
Warn
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides bash commands for the agent to execute in a development environment, specifically
php artisan remote "docs:import --repo=spatie/{repo_name}". This command is designed to perform operations on the production spatie.be instance, meaning successful exploitation could impact production infrastructure. - [INDIRECT_PROMPT_INJECTION]: The
{repo_name}parameter represents a significant attack surface for indirect prompt injection and command injection. - Ingestion points: The variable
{repo_name}is provided by the user or extracted from external context like PR titles in theSKILL.mdinstructions. - Boundary markers: None. The placeholder is interpolated directly into a bash command string without delimiters or warnings to the agent.
- Capability inventory: The skill utilizes local shell access (
cd,&&), production-reaching Artisan commands (php artisan remote), and network tools (curl). - Sanitization: None. The instructions do not define allowed characters or validation steps for the repository name, allowing an attacker to escape the command string using characters like
;,&, or|to execute arbitrary commands locally or on the remote server.
Audit Metadata