The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided information such as target objects, innovation goals, and context for analysis. This creates a surface for indirect prompt injection if the processed data contains hidden instructions.
Ingestion points: Inputs are accepted via target_object, innovation_goal, context, and constraints as defined in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or specific markers to isolate untrusted user data from the core system prompt.
Capability inventory: The skill is restricted to generating text analysis and innovation ideas. No capabilities for network operations, file system writing, or subprocess execution were found in any files.
Sanitization: There are no explicit instructions for the agent to sanitize, escape, or validate the provided input data before processing.
[EXTERNAL_DOWNLOADS]: The reference files contain links to official corporate documentation, academic research, and published books used as educational material.
Evidence: Links include official resources from Google Books, the University of Northern Iowa (scholarworks.uni.edu), McDonald's, the SEC, Starbucks, Apple, Adidas, IKEA, and MUJI.
Context: These references are used for case study analysis and methodology background, which is consistent with the primary purpose of an innovation skill.

scamper