The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted blockchain data (source code, transactions, metadata) which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: RPC endpoints, blockchain explorers, and contract source files as defined in SKILL.md and transaction-forensics.md.
Capability inventory: Execution of cast, forge, solc, slither, and mythril which process these data points.
Boundary markers: The instructions do not define specific delimiters or validation logic for external data.
[COMMAND_EXECUTION]: The skill provides instructions for executing various command-line tools for blockchain analysis.
Evidence: Integration with Foundry suite, Solidity compiler, and security scanners like Slither and Mythril as described in the SKILL.md Tool routing section.

blockchain-ctf