whats-next
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes standard developer tools including
git,gh(GitHub CLI),ls,readlink, andheadto gather project context. These commands are used as intended for local development workflows. - [DATA_EXPOSURE_&_EXFILTRATION]: The skill accesses local project metadata and GitHub issues. This data is used exclusively to generate suggestions for the user and is not transmitted to unauthorized external domains. Access to the GitHub API via the
ghtool is standard for this functionality. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill reads external data from GitHub issue/PR titles and the contents of local markdown files in the project's handoff directory.
- Boundary markers: No explicit delimiters are used to separate untrusted data from instructions, but the usage is confined to providing suggestions for user review.
- Capability inventory: The skill is limited to read-only metadata gathering and shell utilities (
git,gh,ls,head). - Sanitization: No explicit sanitization of issue titles or file content is performed. Given that the output is presented as suggestions for the user to select, the risk is minimal.
Audit Metadata