The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the standard Python library requests for HTTP operations and the vendor-scoped Node.js package @autumnsgrove/lattice for structured error handling and UI feedback in web applications.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external REST APIs, such as GitHub and OpenWeather. Processing unvalidated content from external third-party services represents a potential attack surface for indirect prompt injection.
Ingestion points: api_request and fetch_all_pages functions in SKILL.md retrieve external JSON data.
Boundary markers: The provided code snippets do not include explicit boundary markers or instructions to ignore embedded prompts in the fetched data.
Capability inventory: The skill uses requests.get for network access and json.load for file reading in SKILL.md.
Sanitization: The snippets show basic JSON parsing but do not include specific sanitization or validation of the content returned by the external APIs.

api-integration