The Agent Skills Directory

[COMMAND_EXECUTION]: The script stackspot.ts employs PostConditionMode.Allow for all transaction-signing operations, including join-pot, start-pot, claim-rewards, and cancel-pot. This setting disables the blockchain's native safety checks (post-conditions), which are intended to prevent a contract from transferring more assets than the user explicitly authorized.
[INDIRECT_PROMPT_INJECTION]: The skill fetches and processes data from external blockchain contracts via the get-pot-state and list-pots commands. This creates a potential attack surface where malicious data stored on-chain could influence the agent's behavior.
Ingestion points: Read-only contract calls in stackspot.ts (callPotReadOnly) that retrieve configs, poolConfig, and details fields.
Boundary markers: None identified in the prompt instructions or implementation to distinguish between data and potential instructions.
Capability inventory: The skill has the capability to execute transactions that move user funds (callContract).
Sanitization: No validation or sanitization is performed on the data retrieved from the blockchain before it is processed or presented to the agent.

stackspot