ralph-tui-create-github-issues
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted PRD data which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: Processes PRD content provided as markdown or text as defined in SKILL.md.
- Boundary markers: Lacks explicit delimiters for the full PRD input, though it provides logic for extracting specific sections like "Quality Gates".
- Capability inventory: Utilizes
ghCLI for issue creation and GraphQL API calls, and invokesbunfor linting and type checking. - Sanitization: Explicitly requires the use of single-quoted HEREDOC delimiters (
<<'EOF') in generated commands to prevent shell interpretation of content from the PRD. - [COMMAND_EXECUTION]: Generates and executes commands using the GitHub CLI (
gh) and thebunruntime. These operations are core to the skill's functionality for issue management and task verification. - [DYNAMIC_EXECUTION]: Dynamically assembles shell scripts and GraphQL queries based on the contents of the PRD. The skill mitigates risks by instructing the agent to use safe heredoc syntax.
Audit Metadata