optimizing-prompts
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill's reference documentation (references/role.md) contains mentions of jailbreak patterns like 'DAN' and 'AI without restrictions'. These are documented as 'Anti-patterns' to be avoided or flagged during the optimization process. They do not constitute an attempt to subvert the agent's own safety guidelines.
- [DATA_EXFILTRATION]: The skill includes a constraint to stop and surface an issue if a prompt contains secrets, credentials, or PII. It does not perform any unauthorized network requests or sensitive file access.
- [COMMAND_EXECUTION]: The workflow involves reading local files via the
$ARGUMENTSpath, which is consistent with its stated purpose of optimizing prompt files. No arbitrary command execution or privilege escalation was detected. - [METADATA_POISONING]: The skill's metadata and descriptions accurately reflect its functionality as a prompt engineering assistant.
Audit Metadata