agent-cpp-expert
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided data through the
$ARGUMENTSplaceholder. When combined with systems programming capabilities like file I/O and networking, this establishes a potential surface for indirect prompt injection where malicious instructions in input could attempt to influence the agent's output. However, this configuration is inherent to its primary purpose as a C++ development subagent. - Ingestion points:
$ARGUMENTSin SKILL.md. - Boundary markers: None defined to delimit user input from instructions.
- Capability inventory: The skill claims expertise in systems-level operations including File I/O (mmap, async I/O), Networking (sockets), and Dynamic Loading (dlopen), which implies the agent is equipped to generate and manipulate code for these tasks.
- Sanitization: The skill does not specify filtering or validation for the interpolated input data.
Audit Metadata