kiro-spec-requirements
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows secure and professional patterns for requirement engineering, utilizing a clear separation of concerns between functional requirements and design implementations.
- [COMMAND_EXECUTION]: The skill uses a positional variable
$1to define the target directory for specification files. This variable is used to construct paths for reading and writing project-specific documentation, which is standard functionality for this type of tool. - [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it processes content from project files (e.g.,
brief.md) and external web research findings. This is mitigated by the skill's structure and the inclusion of a rigorous review process. - Ingestion points: Local project specification files and data retrieved during sub-agent research tasks.
- Boundary markers: Uses explicit instructional blocks and a defined "WHAT, not HOW" litmus test to constrain the generated content.
- Capability inventory: Performs file read/write operations and invokes other local developer tools within the project environment.
- Sanitization: A dedicated review gate (
rules/requirements-review-gate.md) enforces EARS syntax, requires numeric IDs, and performs mechanical checks to remove implementation-specific technology language from the requirements.
Audit Metadata