shredstream
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates downloading software and configuration from reputable Solana ecosystem sources, including Jito Labs' official GitHub repository and Anza's release servers for the Solana CLI.
- [COMMAND_EXECUTION]: The documentation includes standard commands for building software with Cargo, running Docker containers with host networking, and configuring systemd services for long-running processes.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted pre-execution transaction data from Solana validators. Ingestion points: 'scripts/parse_shredstream_entries.py' and 'scripts/rabbitstream_monitor.py'. Boundary markers: Absent (data is parsed as binary/protobuf). Capability inventory: Parsing and logging to the terminal. Sanitization: Absent. This ingest-and-process pattern is a standard requirement for the skill's purpose but represents a minor theoretical attack surface for maliciously crafted transaction data.
Audit Metadata