A2A GOSCE Agents
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| structured-output-agentA | Turn a prompt + a field schema into validated, typed JSON (Instructor over Gemini 2.5 Flash on Vertex AI). |
| traced-llm-proxyA | Proxy Gemini (Vertex AI) completions wrapped in OpenTelemetry trace spans; returns the answer plus the trace/span id. |
| authenticated-multi-llm-agentB | Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion for the verified caller. |
| auth-token-serviceB | Hash passwords with bcrypt and issue/verify JWT session tokens. |
| llm-orchestration-agentC | Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing. |
| llm-observability-orchestrationB | Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing. |
| authenticated-llm-agentA | JWT-gated LLM gateway: authenticate (bcrypt/JWT), then run a LangChain-on-Vertex Gemini completion. Unauthenticated calls are rejected. |
| authenticated-mcp-agentA | JWT-gated LLM gateway: authenticate (bcrypt/JWT), then run a LangChain-on-Vertex Gemini completion. Unauthenticated calls are rejected. |
| gosce-portfolio-routerA | Single entry point for the GOSCE portfolio: routes orchestrators to verified agents by capability, with real example output from the router's own /selftest probes (a quality-assured broker, not a directory). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
Two pairs of tools have identical descriptions (authenticated-llm-agent/authenticated-mcp-agent and llm-observability-orchestration/llm-orchestration-agent), making them nearly indistinguishable. Only the router, auth service, structured output, and traced proxy have clear distinct purposes.
Names use hyphens but no consistent prefix or verb-noun pattern. 'authenticated-llm-agent' vs 'authenticated-mcp-agent' are confusingly similar, and the duplicate pairs exacerbate the inconsistency.
At 9 tools, the count is reasonable for a server focused on LLM agents with authentication and tracing. However, the redundancy makes the effective set smaller, but the raw count is not problematic.
The tool set covers user authentication, LLM calls, and tracing, but has significant gaps: no simple non-authenticated LLM call, no user management, and multiple tools overlap without adding distinct capabilities. The purpose of the server is unclear.