The Agent Skills Directory

[PROMPT_INJECTION]: The skill implements a Retrieval-Augmented Generation (RAG) pattern, creating a surface for indirect prompt injection where data from a corpus could influence agent behavior.
Ingestion points: The corpus variable in SKILL.md and example.py is the entry point for data that is later processed by the agent.
Boundary markers: The signature "context: list[str], question -> answer" used in dspy.ChainOfThought lacks explicit delimiters or instructions to ignore embedded commands within the retrieved context.
Capability inventory: The skill has the capability to read from and write to the local file system using the save and from_saved methods for index persistence.
Sanitization: No sanitization or validation of the retrieved text passages is implemented before they are passed to the reasoning module.
[EXTERNAL_DOWNLOADS]: The skill references the installation of faiss-cpu and the use of sentence-transformers for local embedding tasks. These are well-known and standard libraries within the machine learning ecosystem.

dspy-embedding-retrieval