The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl commands to interact with the X2C Open API. It instructs the agent to execute shell-based network requests using environment variables for authentication (X2C_API_KEY) and endpoint configuration.
[DATA_EXFILTRATION]: The skill includes wallet management functionality, specifically the wallet/withdraw-usdc action. This allows the agent to transfer digital assets to a specified Solana address. This capability poses a risk if the destination address is manipulated via malicious input from an untrusted source.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes user-provided metadata and transaction parameters without clear sanitization or boundary markers.
Ingestion points: Metadata fields such as title and description in the distribution/publish workflow, and the to_address parameter in the wallet/withdraw-usdc action (found in SKILL.md).
Boundary markers: Absent; the instructions do not specify the use of delimiters or 'ignore' instructions for the data being interpolated into API requests.
Capability inventory: Shell execution of curl for API interactions and file uploads to S3 (documented in SKILL.md).
Sanitization: None; the skill does not include steps to validate, escape, or filter external data before it is included in network payloads.

x2c-publish