The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes technical documentation from an external source.
Ingestion points: Technical patent description data is retrieved via the https://tool-gateway.linkfox.com/eureka/descriptionData API, as implemented in scripts/eureka_description.py and references/api.md.
Boundary markers: Absent. The instructions in SKILL.md do not recommend wrapping the retrieved patent text in delimiters or providing warnings to ignore instructions embedded within the technical content.
Capability inventory: The skill includes network read operations and telemetry reporting via the Feedback API; it does not perform file system writes or execute arbitrary system commands.
Sanitization: No sanitization, filtering, or validation of the API-returned content is specified in the documentation or the retrieval script.
[DATA_EXFILTRATION]: The skill performs network operations to vendor-specific domains.
It sends requests to tool-gateway.linkfox.com to fetch data and skill-api.linkfox.com for feedback reporting. Both domains are associated with the skill's author, linkfox-ai.
The skill correctly handles sensitive credentials by reading the LINKFOXAGENT_API_KEY from environment variables rather than hardcoding secrets.

linkfox-eureka-description