nemoclaw-community-maintainer-policies
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data from GitHub Issues and Pull Requests to facilitate triage and labeling tasks. While this represents a potential attack surface for indirect prompt injection, the skill mitigates risk through explicit operational constraints.
- Ingestion points: External content from GitHub Issue and Pull Request titles, bodies, and linked metadata as described in references/triage-instructions.md.
- Boundary markers: The skill lacks explicit input delimiters for external content but establishes an output boundary by mandating a dry-run policy and requiring explicit user authorization for all repository metadata writes as specified in references/triage-instructions.md and references/workflow-policy.md.
- Capability inventory: The skill allows for repository metadata modification, including updating labels, titles, and issue types.
- Sanitization: No specific input sanitization or filtering logic is defined for the external data being processed.
Audit Metadata