Loading ToolBench
A configuration and skills repository for the Kiro CLI platform, containing Lambda handlers, Playwright test templates, validation scripts, and setup automation for AWS-based applications.
This is not an MCP server. The repository contains AWS Lambda handlers and Playwright test fixtures, utility functions for a backend system, not MCP tool definitions. The two 'tools' listed (lambda_handler, authenticate) are not registered as MCP tools with schemas, descriptions, or input/output contracts. They are raw Python/TypeScript functions. No MCP protocol implementation is present. The source code shows a Cognito email handler (Python) and a Playwright auth fixture (TypeScript template), neither of which expose themselves as callable MCP tools. Tool definitions are inferred from function signatures, not explicitly declared.
Playwright test setup fixture that performs programmatic authentication via API (custom backend or Cognito), stores auth tokens and user data in localStorage, and persists the authenticated session state to playwright/.auth/user.json for reuse across all E2E tests.
Cognito CustomEmailSender Lambda handler that decrypts KMS-encrypted verification codes, renders branded HTML/text email templates using Jinja2, and sends emails via AWS SES for Cognito triggers (sign-up, password reset, MFA, admin-created user, account takeover notifications).
Not an MCP server. No MCP protocol implementation detected. Repository contains utility functions (Lambda handler, Playwright fixture), not MCP tool definitions.
No input schemas. Both tools lack JSON Schema definitions for parameters. lambda_handler event/context are documented in comments but not as formal schemas. authenticate request/page parameters are TypeScript types, not JSON Schema.
Tool definitions are inferred from source code, not explicitly registered. No MCP tool registration, no manifest, no server entry point.
Generic/unclear naming. 'lambda_handler' is a function name, not a verb_noun tool name (e.g., 'send_cognito_email'). 'authenticate' is vague, does it create a session, validate credentials, or store tokens?
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 24 | <=2025-11-25 | v2 |
No output schemas documented. lambda_handler returns empty dict {}; authenticate returns nothing visible. LLMs cannot plan downstream calls without knowing what fields are available.