Orchestration platform for MCP (Model Context Protocol) integration testing with multiple lab services (DNS, LDAP, TACACS+, mail, NFS, SSO, etc.) and scenario management
Two tools with minimal schema documentation and generic descriptions. endpoints_list and connections_list both lack parameter descriptions and detailed output schemas. Input schemas are present but sparse. Descriptions are adequate (100-200 chars) but lack actionable context for LLM selection. No error handling guidance, no pagination support despite potentially large result sets, and no indication of how secrets are handled in responses despite the descriptions mentioning dev-mode credential exposure.
Get the protocol-level connection details needed to configure a client or system under test against each lab service: protocol endpoints (SMTP, LDAP/LDAPS, DNS, NFS, TACACS+, RADIUS, NTP, HTTPS/OIDC/SAML, MCP) with host:port, client parameters (LDAP base/bind DNs and OUs, DNS zones, NFS mount options, AAA specifics, SSO issuer/JWKS), and the connection credentials (bind passwords, RADIUS/TACACS+ shared secrets, TacLab lab-user passwords, LabLDAP CA PEM, LabSSO CA PEM, IdP Alice password, tokens). Secrets are revealed only in dev mode. Tokens, passwords, and shared secrets come from dev-credentials.yaml when LAB_DEV_MODE=true; lab-ca, LabSSO CA, and TacLab client PEMs are this lab's generated material and are not catalog-stable. Otherwise each credential's usage explains what it is for and how to obtain it.
List the user-facing web/REST endpoints of every lab service (gateway, DNS, LDAP, TACACS+/RADIUS, mail sink, NFS, ...): names, descriptions, URLs, and how to authenticate. In dev mode the response includes the actual credentials (profile-owned via dev-credentials.yaml when LAB_DEV_MODE=true); otherwise it explains how credentials are obtained.
No input parameter descriptions. connections_list accepts 'service' parameter but provides no description of what values are valid, format, or examples.
Output schemas not documented. Both tools lack explicit documentation of response structure, field types, and what data is returned. LLMs cannot plan downstream operations without knowing response shape.
No error handling guidance. Descriptions mention dev-mode credential exposure but provide no error recovery patterns or guidance for LLMs on what to do if a service is unavailable or credentials cannot be retrieved.
Potential secret leakage in responses. Descriptions explicitly state credentials are returned in dev mode, but no guidance on how the tool prevents secrets from entering LLM context or logs in production.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 54 | 2026-07-28+ | v2 |
No pagination or result limiting. Both tools potentially return large result sets (all lab services and their connection details) with no mention of pagination, limits, or how to handle large responses.