The Agent Skills Directory

[COMMAND_EXECUTION]: The SKILL.md file contains a bash script designed to check for skill updates. This script utilizes common shell utilities including curl, grep, awk, and stat to compare local and remote version strings.
[EXTERNAL_DOWNLOADS]: The version check routine performs a network request to raw.githubusercontent.com to fetch a version manifest from the vendor's repository (okx/plugin-store). This is a read-only operation and does not execute remote code.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests and processes data from an external DeFi gateway (https://defi-tools-proxy.saharaa.info).
Ingestion points: Data enters the context via the tokenTool, poolTool, chainTool, protocolTool, and projectTool API endpoints defined in SKILL.md.
Boundary markers: The instructions do not define specific delimiters for the API response data, though it instructs the agent to summarize the output.
Capability inventory: The skill performs network GET requests and returns text for the agent to analyze.
Sanitization: There are no specific sanitization or escaping protocols described for the external content before it is processed by the agent.

sorin-skill