The Agent Skills Directory

[CREDENTIALS_UNSAFE] (HIGH): The skill manages and accesses sensitive secrets, specifically VERCEL_TOKEN and .env files. The command vercel env pull explicitly downloads remote secrets to the local environment, increasing the risk of credential exposure.
[COMMAND_EXECUTION] (HIGH): The skill's core functionality involves executing shell commands via the vercel CLI. These commands can modify remote infrastructure and local project state.
[EXTERNAL_DOWNLOADS] (LOW): Recommends installing the vercel CLI globally via npm. This finding is downgraded to LOW because Vercel is a verified trusted source.
[PROMPT_INJECTION] (HIGH): The skill is highly vulnerable to indirect prompt injection due to its operational workflow.
Ingestion points: The skill reads and interprets package.json and .env files to determine build commands and project configurations.
Capability inventory: Possesses the ability to execute shell commands, access authentication tokens, and deploy code to live production environments.
Boundary markers: None. There are no delimiters or instructions to ignore malicious content within the processed project files.
Sanitization: No sanitization or validation logic is defined for the data extracted from project files before it influences agent decision-making.

vercel-deploy