agency-technical-writer
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No prompt injection or jailbreak attempts were detected. The instructions follow standard instructional patterns for defining a technical writing role and persona.
- [DATA_EXPOSURE]: No hardcoded credentials or sensitive file paths were found. The skill includes documentation templates that use standard placeholders for API keys and configuration (e.g., 'YOUR_APP_ID', 'YOUR_SEARCH_API_KEY', and 'https://app.example.com').
- [REMOTE_CODE_EXECUTION]: The skill does not perform remote code execution or download scripts. While it provides documentation for package installation (e.g., 'npm install'), these are code block examples for the user, not instructions for the agent to execute.
- [OBFUSCATION]: No obfuscation techniques such as Base64 encoding, zero-width characters, or homoglyph attacks were detected. The metadata and description contain legitimate Russian text for technical terms (e.g., 'commits' and 'outdated runbook'), which is appropriate for the stated purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided notes and existing documentation to generate structured technical content. While this represents a data ingestion surface, the skill does not have access to dangerous tools (like shell execution or network requests) that could be exploited by malicious content within those notes.
Audit Metadata