control-flow-abstraction-generator
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No malicious patterns, command executions, or network operations were detected in the skill's instructions or metadata. The content is strictly limited to logic for analyzing code structures and representing them as graphs.
- [NO_CODE]: The skill consists entirely of markdown documentation and reference patterns for the AI to follow. It does not include any scripts, binaries, or configuration files that could execute code on the host system.
- [PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided source code. While this constitutes an attack surface for indirect prompt injection (where instructions are embedded in the code being analyzed), the skill defines no actions beyond textual analysis and formatting (DOT/JSON), providing no mechanism for an injected prompt to perform unauthorized operations.
Audit Metadata