The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes AI-generated documentation and answers from a remote server, creating a surface for malicious content to influence the agent.
Ingestion points: Data enters the agent's context through the output of the read-wiki-contents and ask-question tools in scripts/deepwiki.ts.
Boundary markers: The skill does not use explicit boundary markers or delimiters to isolate untrusted remote content from the agent's instructions.
Capability inventory: The skill performs network operations to mcp.deepwiki.com using the Bun runtime; it does not perform file writes or local subprocess execution.
Sanitization: There is no evidence of sanitization or filtering of the remote service's output before it is presented to the agent.
[DATA_EXFILTRATION]: The skill performs network operations to a domain that is not on the standard whitelist.
Evidence: The CLI script scripts/deepwiki.ts is configured to connect to https://mcp.deepwiki.com/mcp to retrieve repository documentation and metadata.

deep-wiki