The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructs the agent to store sensitive API keys in the credentials/ directory, specifically using files named after a USER_ID. This directory is considered a sensitive path, and accessing files based on user-provided identifiers without explicit validation could lead to path traversal or unauthorized credential access.
[COMMAND_EXECUTION]: All API interactions, including wallet operations like claiming rewards and withdrawing USDC, are performed via curl system commands. Executing shell commands for financial transactions increases the risk of command injection if parameters like addresses or amounts are not properly sanitized.
[DATA_EXFILTRATION]: The X2C_API_KEY is transmitted to a remote endpoint specified by the X2C_API_BASE_URL environment variable. If this variable is pointed to an untrusted domain, sensitive authentication tokens would be exfiltrated to that destination.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (Category 8).
Ingestion points: External data enters the context through video titles, descriptions, and URLs provided during the distribution workflow (SKILL.md).
Boundary markers: The instructions do not include boundary markers or delimiters to isolate untrusted data from the agent's command instructions.
Capability inventory: The skill possesses network and command execution capabilities via curl (SKILL.md).
Sanitization: There is no evidence of sanitization or validation of the ingested strings before they are incorporated into API request payloads.

storyclaw-x2c-publish