The Agent Skills Directory

[METADATA_POISONING]: The skill's YAML frontmatter identifies the author as 'uniswap', which is inconsistent with the registered author identity 'fuzzystodd'. This discrepancy in metadata suggests impersonation of a well-known organization.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from a leader wallet's on-chain transactions and logs, which are external and potentially attacker-controllable inputs.
Ingestion points: The skill uses viem-integration to fetch transactions and Swap logs from a leader wallet as described in Step 2.
Boundary markers: There are no explicit delimiters or specific 'ignore instructions' warnings defined in the workflow for the incoming blockchain data logs.
Capability inventory: The skill utilizes powerful capabilities including Bash(npm:*), Bash(npx:*), Bash(curl:*), WebFetch, and performs delegated swap execution through swap-integration.
Sanitization: The skill implements regex-based input validation for addresses (^0x[a-fA-F0-9]{40}$) and amounts, and explicitly rejects inputs containing shell metacharacters as detailed in the Input Validation section.

copy-trade