The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the user or agent to run shell commands via npx to execute the @cloudbase/cloudbase-mcp server and interact with the mcporter tool manager. It also provides instructions for modifying the local mcporter.json configuration file to register the server.\n- [EXTERNAL_DOWNLOADS]: The instructions involve downloading and executing the @cloudbase/cloudbase-mcp package from the npm registry. It also provides links to official source code repositories on GitHub and cnb.cool, which are managed by the verified vendor (tencentcloudbase).\n- [INDIRECT_PROMPT_INJECTION]: The workflow for calling tools and reading schemas introduces a surface where external data is ingested into the agent context.\n
Ingestion points: Data returned from tools via mcporter call and tool definitions from mcporter describe as referenced in SKILL.md.\n
Boundary markers: The instructions do not specify the use of delimiters or warnings to the agent to ignore potentially malicious instructions embedded in tool outputs.\n
Capability inventory: The skill allows the agent to execute npx and other shell commands within the host environment.\n
Sanitization: There is no mention of filtering, escaping, or validating tool outputs before they are processed by the agent.

cloudbase-mcp