kiro-spec-quick
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
mkdir -pcommand to create project-specific directories for storing generated specifications within the local.kiro/specs/directory structure. - [PROMPT_INJECTION]: The skill processes untrusted data, creating an indirect prompt injection surface. 1. Ingestion points: User arguments ($ARGUMENTS) and the content of local
.kiro/specs/*/brief.mdfiles. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when interpolating external content into prompts or files. 3. Capability inventory: The skill can create directories, write specification files (JSON and Markdown), and invoke downstream tools ($kiro-spec-requirements, $kiro-spec-design, $kiro-spec-tasks). 4. Sanitization: The skill converts descriptions to kebab-case before using them as directory names, which helps mitigate directory traversal risks. - [SAFE]: No external network exfiltration, remote script downloads, or access to sensitive system files (e.g., credentials or SSH keys) was found. The skill's actions align with its described purpose of specification orchestration.
Audit Metadata