dev-scope
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses local file system tools such as Read, Grep, Glob, and Bash solely for surveying project structure and identifying manifest files. These operations are consistent with its purpose of analyzing existing codebases.
- [SAFE]: Strong guardrails are enforced in the instructions, explicitly prohibiting the agent from writing code, scaffolding projects, or installing packages. This limits the agent's capability to only performing documentation tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it reads existing source code to infer project scope. However, this risk is mitigated by a mandatory human-in-the-loop approval step and the lack of any execution capabilities for the ingested data.
- [DATA_EXFILTRATION]: The skill environment lacks network-related tools in the allowed-tools configuration, and the instructions focus entirely on local file operations, preventing any transmission of sensitive data to external servers.
Audit Metadata