agent-evaluation
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill contains examples in
SKILL.mdusingsubprocess.runto execute commands (e.g.,pytest) on agent-generated code for evaluation purposes.\n- [REMOTE_CODE_EXECUTION]: The instructions inSKILL.mdinclude a coding example where theexec()function is used to run arbitrary code strings produced by an agent.\n- [PROMPT_INJECTION]: The skill describes processing untrusted agent data (outcomes, transcripts) which presents an indirect prompt injection surface.\n - Ingestion points: Agent outcomes (
outcome[\"code\"],outcome[\"report\"]) and action transcripts inSKILL.md.\n - Boundary markers: Not explicitly defined in simplified code examples.\n
- Capability inventory: Execution of shell commands and dynamic Python code via
execandsubprocessinSKILL.md.\n - Sanitization: The skill explicitly recommends Docker-based environment isolation for running agent code in
SKILL.md.\n- [SAFE]: References to external benchmarks and research (e.g., Anthropic, SWE-bench, WebArena) point to trusted and well-known official sources.
Audit Metadata