The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the doordash-pp-cli tool using npx from the @mvanhorn/printing-press-library package or via go install from the vendor's GitHub repository.
[DATA_EXFILTRATION]: The CLI tool documentation describes a --deliver webhook:<url> flag, which enables the redirection of command output (including potentially sensitive account or order data) to an arbitrary external network endpoint.
[CREDENTIALS_UNSAFE]: The "Auth Setup" section recommends reading authentication cookies from a file path (/secure/path/cookie-header.txt), which is a sensitive location for credential storage.
[PROMPT_INJECTION]: The skill processes untrusted data from DoorDash (such as search results and menu items) that could contain malicious instructions designed to manipulate agent behavior.
Ingestion points: Data retrieved from DoorDash via the search, menu, item-options, and recent-orders commands described in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or isolate embedded content within the data returned by these commands.
Capability inventory: The skill utilizes the "Read Bash" tool to execute the doordash-pp-cli, providing capabilities for both read operations and account mutations (cart management).
Sanitization: No evidence of sanitization, filtering, or validation is provided for the external content processed by the agent.

pp-doordash