qa
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
gh issue createCLI command to file bug reports on GitHub. This is the primary intended function of the skill and uses standard tooling. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from two sources to generate its output: conversational user input and files within the codebase (e.g.,
UBIQUITOUS_LANGUAGE.md). - Ingestion points: User descriptions of bugs in Step 1 and codebase exploration in Step 2 of
SKILL.md. - Boundary markers: The skill uses structured templates for issues but lacks explicit instructions to treat user-provided data or codebase content as untrusted strings, which could allow embedded instructions to influence the agent.
- Capability inventory: The skill has the capability to execute shell commands (
gh issue create) and spawn subagents (subagent_type=Explore) to read the local filesystem. - Sanitization: There is no mention of sanitizing, escaping, or validating the content retrieved from the codebase or provided by the user before it is interpolated into the issue templates.
Audit Metadata