aicr-release-notes
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
gh(GitHub CLI) tool and a local scripttools/changelogto gather data for release notes. - Evidence: Found in Step 1 of the procedure where
tools/changelogandgh release vieware invoked. - [INDIRECT_PROMPT_INJECTION]: The skill ingests commit messages and previous release bodies, which are external data sources that could contain malicious instructions designed to influence the agent's output.
- Ingestion points: Reads output from
tools/changelog(commit messages) andgh release view(previous release content) as described in Procedure Step 1. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing the external content.
- Capability inventory: The skill writes a Markdown draft to a local temporary file path (
$TMPDIR/aicr-release-notes.md) as specified in Step 5. - Sanitization: There is no evidence of sanitization or filtering applied to the external content before it is processed or written to the file.
- [EXTERNAL_DOWNLOADS]: The skill fetches content from GitHub's official services using the GitHub CLI.
- Evidence: The skill uses
gh release viewto mirror the style of previous releases from the NVIDIA/aicr repository.
Audit Metadata