dflow

Warn

Audited by Snyk on Apr 19, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and act on live, public third‑party data — e.g., DFlow WebSockets (wss://prediction-markets-api.dflow.net) and DFlow Metadata/Trade APIs (pond.dflow.net) as well as Helius DAS (which includes off-chain Arweave/IPFS metadata) — and this data (market/event metadata and real-time feeds) is untrusted/user-provided and directly influences autonomous trading decisions, satisfying the criteria for indirect prompt injection risk.

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

  • Potentially malicious external URL detected (high risk: 0.90). Flagged https://cli.dflow.net because the skill's installation step explicitly runs "curl -fsS https://cli.dflow.net | sh", which fetches and immediately executes remote code at runtime and the DFlow Agent CLI installed from that URL is presented as a required dependency for agent/CLI workflows.

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

  • Direct money access detected (high risk: 1.00). The skill is explicitly designed to execute cryptocurrency trades and submit signed transactions. It documents DFlow trading APIs for spot swaps and prediction-market trades, the DFlow Agent CLI that “handles wallet management, transaction signing, and execution — agents go from prompt to trade in a single command,” and Helius Sender for direct transaction submission/fee optimization. It instructs to “ALWAYS submit DFlow transactions via Helius Sender,” covers “swap tokens on Solana,” “execute trades,” “submit raw transactions,” and wallet/agent configuration for autonomous trading. These are specific crypto/blockchain transaction and wallet APIs intended to move funds, satisfying the Direct Financial Execution criteria.

Issues (3)

W011
MEDIUM

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

W012
MEDIUM

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

W009
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 19, 2026, 07:10 PM
Issues
3