nemoclaw-community-contributor-examples
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and process data from external sources within the repository, such as example README files and pull request metadata. Ingestion points: example READMEs, deployment documentation, and GitHub issue/PR metadata (SKILL.md, references/restructure-checklist.md). Boundary markers: The instructions explicitly require a human contributor to review and correct any agent-generated drafts before they are submitted (SKILL.md). Capability inventory: The skill involves moving files via Git, updating repository links, and editing documentation (references/restructure-checklist.md). Sanitization: The skill contains explicit guidelines to sanitize drafts, remove secrets/internal links, and use placeholders for private environment names (SKILL.md, references/example-taxonomy.md).
Audit Metadata