make-tracking-issue
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (
gh) to perform read and write operations on repository issues, including searching, creating, and editing issues as part of its core functionality. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it gathers information from external sources like repository code, documentation, and existing GitHub issues to generate its output.
- Ingestion points: The workflow includes inspecting local code, package metadata, documentation, and searching existing issues in the repository via the
ghtool. - Boundary markers: The instructions do not define specific delimiters to isolate external data or instruct the agent to ignore embedded commands within that data.
- Capability inventory: The agent uses the
ghCLI tool to read and write GitHub issues and possesses file-reading capabilities for local repository inspection. - Sanitization: No explicit sanitization or filtering mechanisms are described for the data retrieved from external files or GitHub before it is incorporated into the issue drafting process.
Audit Metadata