lgpd-legacy-retrofit
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs legitimate diagnostic operations. It inspects repository structure, configuration templates (e.g., .env.example, Info.plist), and application logs to detect PII leaks as part of an LGPD privacy audit. No sensitive credentials (such as active .env files or SSH keys) are targeted for reading or exfiltration.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project files and logs, creating an attack surface for indirect prompt injection. The risk is minimized by the skill's specific focus on fact-based reporting and the lack of executable capabilities.
- Ingestion points: Project source code, database models (prisma/schema.prisma), dependency manifests (package.json), platform-specific configurations (Info.plist, AndroidManifest.xml), and application log files.
- Boundary markers: No explicit instructions to ignore embedded prompts are provided in the workflow.
- Capability inventory: Limited to file system reads and writing local markdown files within the .lgpd/ directory. No network, subprocess, or shell execution tools are utilized.
- Sanitization: There is no specified sanitization or validation of the ingested repository data.
Audit Metadata