bdd-step-library-curator
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security vulnerabilities or malicious patterns were detected. The skill functions as a local development tool for BDD project maintenance.
- [COMMAND_EXECUTION]: The skill employs standard local commands such as
grep,sed, andgitto analyze the contents of the user's BDD project. These operations are scoped to common source code directories and are used for pattern extraction as described in the documentation. - [DATA_EXFILTRATION]: The skill does not contain any network-facing code, such as
curlorfetch, nor does it attempt to access sensitive system directories or environment variables. All data processing remains local to the execution environment. - [REMOTE_CODE_EXECUTION]: All scripts (Python and Bash) are provided directly within the skill's reference files as plain text. There are no mechanisms to download or execute code from external or unverified remote sources.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted data from the project's source files to perform analysis, it uses static regex-based normalization and generates human-readable reports. The file-writing capabilities (Step 3 and 4) are presented as guided refactoring steps for the agent to perform on specific project files, representing the primary intended purpose of the tool.
Audit Metadata