nemo-relay-use-context-isolation
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists exclusively of documentation (SKILL.md) and evaluation test cases (evals/evals.json). There are no executable scripts, binaries, or automated tool configurations that could present a security risk during agent execution.
- [SAFE]: The documentation focuses on software engineering patterns such as thread-local storage, context variables, and goroutine-safe stacks. It does not instruct the agent to perform sensitive operations, access external network resources, or bypass security boundaries.
- [COMMAND_EXECUTION]: While the skill mentions various CLI-like function names (e.g.,
create_scope_stack(),set_thread_scope_stack()), these are documented as API calls within specific programming libraries rather than shell commands to be executed on the host system. - [PROMPT_INJECTION]: The content does not contain instructions that attempt to override agent safety guidelines, extract system prompts, or induce unrestricted behavior modes.
Audit Metadata