create-ol-github-issue
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data to populate GitHub issue fields, which are then used as arguments in a shell command. This creates a potential surface for indirect prompt injection or command injection if the agent does not properly escape the input.
- Ingestion points: The skill explicitly requests "Repository", "Title", and "Body details" from the user (Step 1), and encourages referencing external source material like logs or error messages.
- Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present when handling user input.
- Capability inventory: The skill uses the
gh issue createcommand to perform its primary action (Step 3). - Sanitization: While the instructions suggest "tighten wording and fix formatting" and stripping HTML comments, there is no guidance on escaping special characters (e.g., backticks, dollar signs, or quotes) that could be used to manipulate the final shell command.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the GitHub CLI (
gh) with parameters derived from user input to create issues. This involves direct interaction with the shell, which is appropriate for the skill's purpose but carries inherent risks if the input is malicious or malformed.
Audit Metadata