The Agent Skills Directory

Official Tool Installation: The skill installs the Stripe CLI and its Projects plugin through verified channels such as Homebrew and the vendor's own plugin manager. These downloads target official repositories and are routine for establishing the required environment.
Dynamic Component Loading: Upon initialization, the skill deploys a secondary component (stripe-projects-cli) to a local directory (.claude/skills/stripe-projects-cli). This is a functional mechanism to extend the agent's capabilities using vendor-authored instructions provided at runtime.
Sensitive Data Handling: The instructions manage project environment variables and API tokens. The skill aligns with security best practices by explicitly directing the agent not to display secret values or print full credential contents in the user interface, only showing variable names.
Indirect Prompt Injection Surface: The skill ingests data from external stripe projects search and stripe projects catalog commands (defined in SKILL.md). There are no specific boundary markers or sanitization protocols mentioned for this ingestion. While the skill possesses capabilities such as shell command execution (Bash) and tool invocation (Skill), the risk is minimal as the data source is the vendor's own catalog service.

stripe-projects