senior-fullstack
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [SAFE]: The skill operates entirely through local Python scripts (
project_scaffolder.pyandcode_quality_analyzer.py) that perform file system operations typical for software development tools. No network exfiltration or malicious code execution patterns were detected within the skill's own implementation. - [COMMAND_EXECUTION]: The skill instructions involve executing local Python scripts and standard package managers (
npm,pip). These operations are necessary for the skill's primary purpose of project scaffolding and auditing and utilize standard tools in a legitimate manner. - [INDIRECT_PROMPT_INJECTION]: The
code_quality_analyzer.pytool reads user-provided source files to perform static analysis. While this ingestion of untrusted data represents a potential attack surface, the tool uses fixed regular expressions for scanning and does not dynamically execute the analyzed content, mitigating risks to the agent environment. - [CREDENTIALS_UNSAFE]: The scaffolding tool generates environment templates with placeholder secrets (e.g., 'your-secret-here', 'password'). The documentation explicitly flags using these default values in production as an anti-pattern and provides clear guidance on secure secret management.
Audit Metadata