polymarket-whale-trades

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted trade data from the Polymarket exchange via the 0xinsider API, which could be used to deliver hidden instructions to the agent through market event descriptions or metadata.
  • Ingestion points: Data is ingested through multiple API endpoints described in SKILL.md, including get_whale_trades, get_whale_trade, and get_event_replay_since.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the skill instructions.
  • Capability inventory: The skill utilizes network read operations via curl and MCP tools; no capabilities for file system modification or arbitrary code execution were detected.
  • Sanitization: The instructions do not define any validation or sanitization procedures for the data retrieved from the external API.
  • [EXTERNAL_DOWNLOADS]: The skill documents the use of an external API sandbox for retrieving trade information.
  • Evidence: The skill provides a curl example targeting https://0xinsider.com/sandbox/api/v1/whale-trades, which is a resource hosted on the vendor's primary domain.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:49 PM
Security Audit — agent-trust-hub — polymarket-whale-trades