bof-dev
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to guide the agent in generating and auditing low-level C and C++ source code for Beacon Object Files (BOFs), which creates a surface where instructions embedded in processed code could influence agent behavior.
- Ingestion points: The skill ingests user-provided C/C++ logic and requirements to apply BOF-specific constraints (e.g., DFR patterns) as defined in
SKILL.mdand referenced files. - Boundary markers: The skill establishes 'Strict Rules' in
SKILL.mdto constrain output to valid BOF constructs, but it lacks explicit instructions to disregard embedded natural language prompts within comments or strings in user code. - Capability inventory: The skill provides instructions for utilizing the
x86_64-w64-mingw32-gcccompiler and interacting with system-level Win32 APIs through thebeacon.hinterface. - Sanitization: There are no explicit sanitization mechanisms or validation steps defined to filter instruction-carrying strings from the processed source code before it is integrated into the agent's context.
Audit Metadata