ln-112-project-core-creator

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs project-wide scans and uses search patterns to auto-discover technologies and architectural patterns. These operations are localized to the project environment and support its core documentation mission.
  • [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection by ingesting data from LEGACY_CONTENT and contextStore (ingestion points in SKILL.md) for interpolation into templates. Mitigation is provided through boundary markers like <!-- NO_CODE_EXAMPLES --> (templates), limited capabilities (localized file writing and grep), and automated validation in Phase 3 (SKILL.md).
  • [SAFE]: No evidence of credential theft, malicious persistence, or unauthorized data exfiltration was found. The skill operates within its defined scope to produce static documentation files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 08:32 AM