backend-type-bridger
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions and examples include commands for fetching an OpenAPI schema from a local development server (
curl -fsS http://127.0.0.1:8090/openapi.json -o openapi.json). While this involves a network download, it is restricted tolocalhost(127.0.0.1), which is a whitelisted domain/interface for development workflows. The skill explicitly warns against passing remote URLs into the sync script to prevent SSRF or untrusted data ingestion. - [REMOTE_CODE_EXECUTION]: Automated scans flagged the
curlcommand to127.0.0.1. However, this is not remote code execution; it is a standard development task to fetch a schema from a local running service to a file. The downloadedopenapi.jsonis parsed as JSON, not executed as code. - [DATA_EXFILTRATION]: No data exfiltration patterns were detected. The scripts read local configuration files (
firestore.rules,openapi.json) and write generated TypeScript interfaces to the local filesystem (src/types/api.ts). - [SAFE]: The skill implements strict path resolution (
resolve_openapi_path) to ensure all file operations remain within the project root. It also limits the amount of metadata emitted to the agent context to prevent context stuffing.
Audit Metadata