The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation guides the agent to perform network requests to official Polymarket API domains including gamma-api.polymarket.com, clob.polymarket.com, and data-api.polymarket.com. These are well-known service endpoints for retrieving public market data and do not involve authentication or sensitive data transmission.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted market data retrieved from external APIs (market questions, descriptions, outcomes). While this presents a surface for indirect prompt injection, the risk is mitigated by the read-only nature of the skill and the standard data formats (JSON) used for ingestion. The skill lacks capabilities for command execution or file writing that would escalate this risk beyond a low level.
Ingestion points: Data enters the context via gamma-api.polymarket.com, clob.polymarket.com, and data-api.polymarket.com (SKILL.md, references/api-endpoints.md).
Boundary markers: None explicitly defined in instructions, but the usage of structured JSON responses provides inherent data separation.
Capability inventory: The skill only performs read-only network operations (GET requests). It does not have file-write or shell-execution capabilities.
Sanitization: Instructions specify formatting prices as percentages and presenting question strings, but do not specify input validation or sanitization for market metadata strings.

polymarket