hotfix
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted user input through the $ARGUMENTS variable, which is then passed to internal review and triage agents. This creates a surface for indirect prompt injection if the user provides a malicious bug description. Ingestion points: User-provided bug descriptions, issue links, or error messages are captured in the bug_description section of SKILL.md. Boundary markers: The input is encapsulated within <bug_description> tags to demarcate it from system instructions. Capability inventory: The skill can invoke multiple agents and execute shell commands for git operations and file cleanup. Sanitization: No explicit validation, filtering, or sanitization logic is applied to the user input before it is utilized in agent tasks.
- [COMMAND_EXECUTION]: The skill executes local shell commands for repository management and temporary file cleanup. Evidence: The skill performs git operations (git rev-parse, git checkout, git commit, git push) and directory removal (rm -rf docs/hotfix-review/).
Audit Metadata