The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process data from Mercado Libre, including product descriptions, user reviews, and buyer questions, which are considered untrusted external inputs.
Ingestion points: API responses from https://api.scavio.dev containing external, untrusted text fields such as description, reviews[].text, and questions[].text in SKILL.md examples.
Boundary markers: The instructions lack explicit delimiters or guidance for the agent to ignore potentially malicious instructions embedded within the fetched data.
Capability inventory: The skill utilizes network operations via requests and the scavio library to fetch data.
Sanitization: No specific sanitization or filtering of the external text content is provided in the skill instructions.
[EXTERNAL_DOWNLOADS]: The documentation guides the installation of external dependencies and tools.
Python dependency: scavio==0.21.0 via pip.
Node.js dependency: scavio@0.21.0 and @scavio/mcp-server@0.19.1 via npm.
These resources are identified as vendor-owned by scavio-ai.
[COMMAND_EXECUTION]: The documentation provides instructions to execute a Model Context Protocol (MCP) server.
Evidence: npx -y @scavio/mcp-server@0.19.1 in SKILL.md under the MCP section.

mercadolibre-product-data