The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from the Neon API, such as project names, branch names, and database metadata, which could theoretically contain malicious instructions.
Ingestion points: External data enters the agent context via tools like NEON_RETRIEVE_PROJECTS_LIST, NEON_GET_BRANCHES_FOR_PROJECT, and NEON_FETCH_DATABASE_FOR_BRANCH.
Boundary markers: The skill instructions do not specify the use of delimiters or specific instructions to ignore embedded content in API responses.
Capability inventory: The skill includes capabilities to list, inspect, and retrieve connection URIs for database resources, but does not include arbitrary code execution or file-system writing tools.
Sanitization: No explicit sanitization or filtering of API-returned strings is mentioned in the skill definition.

Neon Automation