codex-issue-digest
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [Command Execution]: The skill uses the
subprocessmodule to interface with the GitHub CLI (gh) and Git. This is necessary to fetch the data required for the digest. The implementation follows security best practices by passing arguments as a list and sanitizing label inputs to mitigate injection risks. - [Processing External Data]: By its nature, the skill ingests content from GitHub issues and comments. While this introduces a surface for indirect prompt injection if an issue contains instructions, the skill provides clear guidance to the agent on how to summarize and cluster this information into a report, which helps maintain the intended behavior.
- [Network Activity]: The skill performs network requests to the GitHub API via the
ghcommand-line tool. These operations are essential for the skill's functionality and are limited to the official GitHub API for the specified repository.
Audit Metadata