The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches and processes user-generated content such as reviews, business descriptions, and AI summaries from Trustpilot via the Scavio API, which creates a surface for indirect prompt injection.
Ingestion points: Data enters the agent's context through the reviews[], description, and ai_summary fields returned by endpoints like /api/v1/trustpilot/business and /api/v1/trustpilot/reviews (found in SKILL.md).
Boundary markers: The skill returns structured JSON data, which provides some structural separation, but the instructions do not explicitly mandate the use of delimiters or provide warnings to the agent to ignore instructions embedded within the review text.
Capability inventory: The skill provides read-only access to Trustpilot data and does not include capabilities for file system modification, local command execution, or other high-privilege operations.
Sanitization: There is no evidence of sanitization or filtering of the external text content to remove potential injection payloads before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill documentation recommends the installation of external packages and the use of an MCP server to facilitate API interaction.
Evidence: The instructions suggest pip install scavio==0.19.1, npm install scavio@0.19.0, and running npx -y @scavio/mcp-server@0.17.1 (found in SKILL.md).
Context: These resources are provided by the skill's author (scavio-ai) and appear to be legitimate tools for accessing their service.

trustpilot-reviews-api