The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the configuration and use of the Patsnap MCP (Model Context Protocol) service, which involves connecting to external data APIs hosted at open.zhihuiya.com for patent intelligence and legal status information.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, including patent abstracts, claims, and web search results from third-party sites. This ingestion represents a potential attack surface for indirect prompt injection if malicious instructions are embedded within the retrieved technical documents or web pages.
Ingestion points: Data enters the context via patsnap_search, patsnap_fetch, and web_search tool outputs.
Boundary markers: The instructions do not define specific delimiters for separating tool-provided technical data from system instructions.
Capability inventory: The skill has the capability to write local files (HTML, Word, Excel/CSV) and perform network-based tool calls.
Sanitization: Employs 'Name Normalization' logic to aggregate entities, but does not specify technical escaping for the generated HTML output content.
[DYNAMIC_EXECUTION]: The skill dynamically generates interactive HTML wizard pages and reports based on data processed at runtime. It also specifies the creation of exportable Word and Excel/CSV files, which involves assembling structured and unstructured data into executable or formatted document containers.

tech-transfer-target-discovery