write-natural-swedish
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFENO_CODEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown instructions and reference files without any executable scripts, binaries, or automated shell commands.
- [EXTERNAL_DOWNLOADS]: The references/resource-map.md file includes links to trusted Swedish linguistic institutions and official media outlets, such as the Swedish Academy and Språkrådet, for terminology verification.
- [PROMPT_INJECTION]: The skill processes untrusted text input via its arguments in SKILL.md, establishing an indirect prompt injection surface. The analysis found boundary markers in SKILL.md directing the agent to preserve factual meaning. A capability inventory across all files confirms there are no subprocess calls, network operations, or file-system writes, which mitigates the risk of malicious input. No input sanitization is specified.
Audit Metadata