nemoclaw-contributor-update-docs
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes local shell commands including
git logandgit showto analyze repository history,make docsto build the documentation, andgh pr createto automate the submission of updates. - [COMMAND_EXECUTION]: During the release preparation phase, the skill executes a repository-specific Python script
scripts/docs-to-skills.pyto convert documentation into user skill definitions. - [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface (Category 8) by ingesting untrusted data from git commit messages and code diffs.
- Ingestion points: Commit messages and diffs are read from git history in Step 1 and Step 3.
- Boundary markers: No explicit boundary delimiters are used to isolate commit data, though Step 5 provides a detailed style guide for the agent's output.
- Capability inventory: The skill possesses capabilities to write to the local file system (
docs/), execute local Python scripts, and open PRs on GitHub. - Sanitization: Content is filtered against a local skip-list (
.docs-skip), but there is no explicit sanitization for malicious instructions embedded within the processed git history.
Audit Metadata