The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from an external news website, creating an attack surface for indirect prompt injection.
Ingestion points: Tools such as expressen_se__read_page, expressen_se__get_sok, and browsing tools listed in SKILL.md fetch content from expressen.se.
Boundary markers: The instructions do not specify any delimiters or safety warnings for the agent when handling the retrieved news content.
Capability inventory: The skill performs network operations to the unbrowse.ai vendor API to execute tasks.
Sanitization: No explicit sanitization or filtering of external content is mentioned.
[EXTERNAL_DOWNLOADS]: The skill relies on remote network operations to provide its core functionality.
The MCP server is hosted at https://unbrowse.ai/mcp/expressen.se.
REST calls are made to https://unbrowse.ai/api/v1/sites/expressen.se/call/....
These resources are owned by the vendor unbrowse-ai.

expressen-se