The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill performs network operations to https://mcp.deepwiki.com/sse. This external endpoint is used to proxy tool calls for fetching documentation and asking questions about repositories.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8). It ingests data from a remote AI service (mcp.deepwiki.com) and presents it to the agent without boundary markers or sanitization. Malicious instructions embedded in the AI-generated documentation could potentially influence the agent's behavior.
Ingestion points: Data enters the agent context via the output of read-wiki-contents and ask-question tools.
Boundary markers: None detected. The script directly prints results from the remote server.
Capability inventory: The script performs network requests via the mcporter runtime but does not appear to have direct file-write or local command execution capabilities beyond the CLI invocation itself.
Sanitization: No sanitization or validation of the remote content is performed before it is output to the user/agent.
[COMMAND_EXECUTION]: The skill requires the use of the bun runtime to execute the deepwiki.ts script. The documentation provides example commands that use absolute local file paths (C:\Users\lyf1143312445\...), which indicates the skill may be configured for a specific local environment.

deep-wiki