The Agent Skills Directory

[CREDENTIALS_UNSAFE]: Hardcoded API key provided for demonstration purposes.\n
Evidence: The file SKILL.md and references/usage-patterns.md contain the literal API key string d573a84c-8e79-4a35-b0c6-427e9ad2478d labeled as a shared demo key for trial usage.\n- [COMMAND_EXECUTION]: Relies on shell command execution of external CLI tools.\n
Evidence: The skill instructs the agent to use uxc to link and execute the okx-mcp-cli command to interact with the OKX API.\n- [INDIRECT_PROMPT_INJECTION]: Vulnerability surface through the processing of untrusted external data.\n
Ingestion points: Untrusted data enters the agent context via API responses from web3.okx.com, specifically token search results, ranking info, and market metadata.\n
Boundary markers: Instructions suggest using JSON output envelopes, but there are no specific delimiters or warnings to ignore embedded instructions within the processed text.\n
Capability inventory: The skill can execute high-impact operations such as dex-okx-dex-swap and dex-okx-dex-approve-transaction.\n
Sanitization: No explicit sanitization or filtering of external content is specified before interpolation into the prompt context.

okx-mcp-skill