blockchain-basics
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes arbitrary user-supplied task descriptions, which presents a standard surface for indirect prompt injection.
- Ingestion points: User task descriptions passed via the primary command interface (referenced in references/GUIDE.md).
- Boundary markers: Absent; no specific instructions are provided to the agent to delimit or ignore instructions within the user input.
- Capability inventory: The skill instructions provide cryptographic primitives and local file system metadata access through the scripts/validate.py validation utility.
- Sanitization: No explicit input sanitization or filtering logic is defined for the task description before it is processed by the agent.
- [SAFE]: All Python code examples in SKILL.md for Merkle proof verification and block hashing utilize standard, reputable libraries like hashlib and struct.
- [SAFE]: The signature verification examples leverage the well-known and trusted eth-account library for secure cryptographic operations.
- [SAFE]: The included validation script (scripts/validate.py) uses yaml.safe_load and performs only non-destructive, read-only checks on the local skill directory structure.
Audit Metadata