agent-observability
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
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| runA | Run the agent-trace CLI with the given subcommand/args; --json is appended automatically so the result is machine-readable JSON. Only these subcommands support --json today: list, replay, inspect, diff, run. Real Run the agent-trace CLI (agent-observability-trace-cli). |
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 1 tool
There is only one tool exposed, so there is no possibility of confusing it with another tool at the MCP level. All subcommands are contained within the single 'run' interface, so no ambiguity exists between distinct tools.
The tool name 'run' is a generic verb without a noun, so it does not follow a consistent verb_noun pattern. Since there is only one tool, there is no inconsistency, but the name is vague and provides little semantic guidance.
A single tool for an entire agent-observability domain is far too thin, forcing all operations through one generic CLI wrapper. This makes the server feel more like a shell escape than a purpose-built MCP service; more granular tools for listing, replaying, and inspecting traces would be appropriate.
The tool can invoke any CLI subcommand, so the underlying functionality may be accessible, but the MCP surface itself is a single generic entry point. Agents cannot discover available operations without reading help output, which is a notable usability gap, though the CLI wrapper does allow core workflows to execute.