The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes several Python scripts, such as blockchain_explorer.py and chain_client.py, which are executed as CLI tools to perform blockchain queries and data processing. These scripts serve as the primary mechanism for the skill's functionality.\n- [EXTERNAL_DOWNLOADS]: To retrieve blockchain data, the skill makes network requests to several well-known and reputable services, including Etherscan, Polygonscan, BSCScan, CoinGecko, and LlamaNodes. These connections are documented and essential for the skill's operation.\n- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it ingests and processes untrusted data directly from public blockchain transactions and smart contract logs. This data is decoded and displayed, potentially allowing an attacker to embed malicious instructions in on-chain data that could influence an agent's logic.\n
Ingestion points: scripts/chain_client.py (via RPC and Explorer API calls to retrieve transaction details and logs).\n
Boundary markers: Absent; there are no explicit delimiters or instructions to ignore content within the retrieved blockchain data.\n
Capability inventory: The skill utilizes the Bash tool to execute its own Python scripts, which have network and data manipulation capabilities.\n
Sanitization: No explicit sanitization or filtering is applied to the retrieved transaction data before it is formatted for output.

exploring-blockchain-data