The Agent Skills Directory

Remote Code Execution (CRITICAL): The skill instructs users to install software using curl -sSL https://canifi.com/skills/stripe/install.sh | bash. Piped remote execution from an untrusted domain allows for full system compromise as the script can execute arbitrary commands with the user's privileges without prior inspection.
External Downloads (HIGH): The installation process fetches scripts from canifi.com, which is not a verified or trusted source. There are no integrity checks (e.g., checksums or signatures) to ensure the scripts have not been tampered with.
Credentials Unsafe (HIGH): The skill prompts users to store SERVICE_EMAIL and SERVICE_PASSWORD in environment variables via canifi-env. These sensitive credentials, while stated to be "local," are readily accessible to the agent or any other script running in the same environment, creating a high risk of credential theft.
Indirect Prompt Injection (LOW): The skill ingests untrusted data from the Stripe dashboard (e.g., transaction notes, customer names).
Ingestion points: Stripe Dashboard (dashboard.stripe.com) via Playwright MCP.
Boundary markers: Absent; the skill does not define delimiters for external data.
Capability inventory: Shell execution (bash) and browser automation (Playwright).
Sanitization: Absent; there is no mention of filtering or escaping data read from the dashboard before processing.

stripe