The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill includes a hardcoded project identifier (OK-ACCESS-PROJECT: 4d156bf0c61130f2692d097ecb68dbe4) within the references/cli-reference.md file, which is used as a header for authenticating requests to the vendor's API.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves and displays data from external APIs that could potentially contain malicious instructions intended to influence the agent's behavior.
Ingestion points: Data is ingested through the onchainos competition list, onchainos competition detail, and onchainos competition rank commands.
Boundary markers: The skill uses strict formatting templates and markers like <MUST> and <NEVER> for the agent, but it does not implement specific delimiters or 'ignore' instructions for the data retrieved from the APIs.
Capability inventory: The skill has the ability to execute shell commands (onchainos), access wallet addresses, and delegate trade executions to other skills.
Sanitization: There is no evidence of explicit sanitization or filtering of the natural language strings returned by the competition APIs before they are rendered to the user.
[COMMAND_EXECUTION]: The skill relies extensively on the onchainos CLI tool to perform blockchain queries, registration, and reward claims, which involves executing system-level commands.

okx-growth-competition