tracking-token-launches
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it fetches untrusted data from external blockchain RPC endpoints and market APIs (e.g., DexScreener, GeckoTerminal) to display token metadata and contract details.
- Ingestion points: External data is ingested through blockchain logs and API responses in
scripts/event_monitor.pyandscripts/token_analyzer.py. - Boundary markers: The instructions do not define specific delimiters to wrap or isolate the external content when presented to the agent, though it is generally structured into tables or JSON.
- Capability inventory: The skill has access to
Bash(restricted to specific Python scripts),Read,Write, andEdittools. - Sanitization: The scripts use basic string decoding with error handling (
errors="ignore") for token symbols and names inscripts/event_monitor.pyandscripts/token_analyzer.py.
Audit Metadata