opus5-prompting-guide
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation and guidance tool for prompt engineering. It provides instructions on how to optimize prompts for specific model behaviors (thinking blocks, effort levels, tool usage).
- [SAFE]: All external URLs referenced in the skill (e.g., platform.claude.com, code.claude.com) belong to Anthropic, which is a well-known and trusted service provider. These links are used appropriately for technical documentation and reference purposes.
- [SAFE]: The skill includes explicit security best practices, such as a section on marking and handling potentially untrusted pasted content using tags (
<pasted_content>) to defend against indirect prompt injection. - [SAFE]: No suspicious command execution, persistence mechanisms, or credential harvesting patterns were detected. The technical details regarding settings (e.g.,
~/.claude/settings.json, environment variables) are documented in the context of project configuration and model behavior rather than unauthorized access.
Audit Metadata