pancakeswap-v3-plugin

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The workflow queries public JSON-RPC endpoints and subgraphs, which return data containing external contract output or subgraph records, but does not monitor or consume unstructured outsider-authored free text feeds without specific target item selection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill includes inline shell commands in SKILL.md that fetch and execute remote scripts and binaries from GitHub (raw.githubusercontent.com and GitHub Releases) via curl at runtime.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 0.70). The skill provides commands to swap tokens and manage concentrated liquidity pools on PancakeSwap V3 across multiple chains, executing direct on-chain financial operations (swap, add-liquidity, remove-liquidity) via wallet contract calls. While the documentation describes confirmation protocols and manual preview steps, these instructions rely entirely on the AI agent following them rather than an independent service- or execution-layer authentication/authorization gate that prevents unauthorized signing.