The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill implements a workflow for fetching and parsing data from external network sources, which introduces a surface area for indirect prompt injection if the ingested data is later processed by an LLM without strict sanitization.
Ingestion points: Data is retrieved via HTTP GET requests in the fetchUser and fetchUsers functions within SKILL.md.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the examples.
Capability inventory: The skill utilizes network access (GET requests) and object instantiation based on external input.
Sanitization: The implementation uses explicit type casting (as Map<String, dynamic>) and Dart's pattern matching in the fromJson factory constructor, providing structural validation of the JSON schema.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill demonstrates standard network operations using the http package. The URLs used (https://api.example.com/users/) are common placeholders and do not represent a threat.
[EXTERNAL_DOWNLOADS]: The skill references the standard Flutter http and foundation packages, which are well-known and trusted components of the Flutter ecosystem.

flutter-implement-json-serialization