skills/helius-labs/core-ai/phantom/Gen Agent Trust Hub

phantom

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes untrusted on-chain data (e.g., transaction descriptions, token metadata) through Helius API calls as described in SKILL.md and references/helius-das.md. While the skill does not explicitly define boundary markers for this data, it provides a strong inventory of capabilities and emphasizes sanitization through server-side signature verification and backend proxying as documented in references/token-gating.md and references/frontend-security.md.\n- [REMOTE_CODE_EXECUTION]: The skill suggests running npx helius-mcp@2.2.0 in SKILL.md to install the Helius MCP server. This involves remote code download and execution, but it is consider safe in this context as it uses the official package provided by the vendor (helius-labs).\n- [COMMAND_EXECUTION]: The skill includes instructions for running various developer CLI tools, such as npx -y create-solana-dapp@latest and helius signup in references/browser-sdk.md and references/helius-onboarding.md. These are standard tools for the intended Solana development workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 01:32 PM