Skills
Navigate
skills/nvidia/nemoclaw/nemoclaw-maintainer-refactor-docs/Gen Agent Trust Hub

nemoclaw-maintainer-refactor-docs

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes repository documentation files that could theoretically contain embedded instructions meant to influence the agent.
  • Ingestion points: The agent is instructed to read content from docs/, fern/, README.md, test, scripts, docs/index.yml, and fern/docs.yml as identified in SKILL.md.
  • Boundary markers: There are no explicit instructions to use delimiters or to disregard instructions found within the source documentation text.
  • Capability inventory: The skill involves writing to the file system (to refactor pages) and executing shell commands via npm run and npx vitest for build validation.
  • Sanitization: No mechanisms for escaping or sanitizing documentation content before interpolation or modification are specified.
  • [COMMAND_EXECUTION]: The skill employs standard developer tools for repository management and validation.
  • Evidence: Uses rg (ripgrep) for content discovery and executes validation scripts via npm run docs:sync-agent-variants, npm run docs, and npx vitest (Step 7 in SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 01:19 PM