emblem-ai-prompt-examples
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation providing text-based prompt templates for users. It does not perform any operations on its own.
- [NO_CODE]: No executable scripts (JavaScript, Python, Shell) or binaries are included in the skill package.
- [PROMPT_INJECTION]: The provided prompt examples do not contain instructions to override AI safety guidelines, disregard system prompts, or bypass ethical constraints. In fact, many examples include explicit instructions to 'do not execute' or 'stop for confirmation'.
- [DATA_EXFILTRATION]: No network-capable commands (e.g., curl, wget) or sensitive file path accesses were found. All crypto addresses and vault IDs mentioned are clearly marked as placeholders or public examples for illustration.
- [INDIRECT_PROMPT_INJECTION]: While the skill includes prompts designed to analyze 'operator-provided sources', it specifically instructs the agent to 'flag what still needs onchain verification' and 'flag unsupported claims', which are defensive patterns against untrusted data.
Audit Metadata