The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to download and execute an installation script for the uxc CLI tool from the author's official GitHub repository (https://raw.githubusercontent.com/holon-run/uxc/main/scripts/install.sh). This is part of the standard setup process for the vendor's tools.
[COMMAND_EXECUTION]: The skill utilizes shell commands to manage tool linking and execution. It uses uxc link to create a local command alias (deepwiki-mcp-cli) and subsequently executes this alias to perform documentation queries. It also includes a validate.sh script for development-time structure verification.
[DATA_EXFILTRATION]: User queries and repository identifiers are sent to the external endpoint mcp.deepwiki.com/mcp. This network communication is the primary function of the skill to retrieve AI-powered answers from the DeepWiki documentation index.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes content from external GitHub wikis and repositories. If these documentation sources contain malicious instructions, they could potentially influence the agent's behavior.
Ingestion points: Content retrieved via read_wiki_contents and ask_question tools (SKILL.md).
Boundary markers: None explicitly defined in the tool prompts to separate external data from instructions.
Capability inventory: The agent has the capability to execute shell commands via the deepwiki-mcp-cli tool.
Sanitization: No explicit sanitization or filtering of the documentation content is performed within the skill definition.

deepwiki-mcp-skill