The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by ingesting and processing untrusted data from Git repositories and Pull Requests to automate Vercel deployments.
Ingestion points: Git repository content, branch names, and Pull Request metadata referenced in the deployment instructions.
Boundary markers: None present; the skill lacks specific delimiters or instructions to ignore potential malicious content embedded within the code being deployed.
Capability inventory: The skill uses Bash(vercel:*), Bash(curl:*), Bash(git:*), and Write tools to perform deployments, status checks, and environment configuration.
Sanitization: There is no evidence of sanitization or validation of the repository-sourced content before it is processed by the deployment tools.
[DATA_EXFILTRATION]: The skill uses curl and the Vercel CLI to transmit project files and configuration data to api.vercel.com. This behavior is consistent with the skill's primary purpose and targets a well-known deployment service.
[COMMAND_EXECUTION]: The skill provides instructions to execute shell commands using vercel, git, and curl to manage the deployment lifecycle, which is standard for CLI-driven deployment workflows.

vercel-deploy-preview