sembr
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is text reformatting following the sembr.org specification. It contains explicit 'MUST' rules to preserve word-for-word integrity and avoid modifying functional regions such as code blocks, YAML/TOML frontmatter, and HTML islands, which effectively prevents accidental corruption of metadata or executable parts of a file.
- [SAFE]: No evidence of data exfiltration, network requests, or sensitive file access was found. The skill does not use any external dependencies, remote scripts, or hardcoded credentials.
- [SAFE]: The instructions do not contain any prompt injection attempts or attempts to bypass agent safety filters. The logic is strictly focused on linguistic structure and line-wrapping rules for document maintainability.
- [SAFE]: While the skill processes user-provided prose (a surface for indirect prompt injection), it lacks instructions that would allow for the execution of code found within that prose. Furthermore, its 'no-touch' rules specifically exclude sensitive areas like frontmatter and code blocks from processing.
Audit Metadata