dod-adherence-review
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a structured audit workflow that verifies repository state using objective artifacts. Its core logic is designed to resist manipulation by explicitly instructing the agent to disregard self-attestations and unverified claims.
- [EXTERNAL_DOWNLOADS]: The documentation references external resources from well-known and trusted organizations, including the Scrum Guide (scrumguides.org), GitHub CLI documentation (cli.github.com), and axe-core accessibility tools (github.com/dequelabs). These are used for instructional purposes and do not involve remote code execution.
- [COMMAND_EXECUTION]: The skill instructs the agent to use standard, read-only command-line tools such as
git diffandgh pr viewto gather evidence for the compliance audit. - [PROMPT_INJECTION]: The skill addresses potential indirect prompt injection by establishing a verification standard that requires tool-generated proof (e.g., LCOV files, GitHub API responses) for every audit point, thereby reducing the impact of malicious instructions that might be embedded in pull request descriptions or checklist files.
Audit Metadata