The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands (curl) to interact with the Denser Retriever REST API for managing knowledge bases and uploading files. It also uses python3 as a utility for parsing and formatting JSON responses returned by the API.
[INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted documents (PDF, DOCX, PPTX, etc.) and imports their content into the agent's context for semantic search and retrieval. This creates a surface where a malicious document could contain hidden instructions intended to influence the agent's behavior.
Ingestion points: External files and text content are ingested through the importFile and importTextContent endpoints and then retrieved via the query endpoint.
Boundary markers: The instructions do not specify any delimiters or boundary markers to isolate retrieved document content from the rest of the agent's prompt.
Capability inventory: The skill has the ability to execute network requests via curl, read local files for upload, and write API responses to the console.
Sanitization: There is no mention of content sanitization or filtering of the document text before it is returned to the agent in search results.

denser-retriever