Spoor
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SPOOR_LOGS_URL | No | URL of the VictoriaLogs backend (e.g. http://localhost:9428) | |
| SPOOR_TRACES_URL | No | URL of the VictoriaTraces backend (e.g. http://localhost:10428) | |
| SPOOR_METRICS_URL | No | URL of the VictoriaMetrics backend (e.g. http://localhost:8428) |
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": true
} |
| logging | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_servicesB | Services seen in the window and which signals (metrics, logs, traces) have data for each. |
| new_errorsA | Error logs compared with a baseline: errors that are new or spiking (messages that differ only by IDs count as one error), services whose error volume rose beyond their traffic, each with counts, first-seen time and one redacted example line, plus services whose errors are steady. |
| trace_diffA | Span names that got slower or started failing compared with a baseline: p95 and error rate before/after, each span's share of the added latency, and an example trace ID. |
| what_changedA | Ranked list of what changed in a service (or all services) across metrics, logs and traces, compared with a baseline. Returns a one-paragraph summary, findings with before/after values, onset times and evidence links, and per-signal coverage. Start here. |
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 4 tools
Each tool maps to a distinct signal scope: list_services inventories services, new_errors is log-specific, trace_diff is trace-specific, and what_changed is the cross-signal ranked entry point. The only soft overlap is that what_changed subsumes findings from new_errors and trace_diff, but its 'start here' framing and ranked-summary output keep the boundary workable.
All names use snake_case casing, but the conventions are mixed: list_services is verb_noun, new_errors is adjective_noun, trace_diff is noun_noun, and what_changed is a question-phrase. Readable and memorable, but there is no predictable verb_noun pattern an agent can rely on.
Four tools for a baseline-diffing observability server is well-scoped: one inventory, one per signal (logs, traces), and one aggregator. Each tool clearly earns its place with no gratuitous or redundant entries.
The surface covers service inventory, log errors, trace regressions, and a unified change ranking, which handles the core diagnostic workflow. Gaps are minor: no metric-specific diff tool (metrics only surface inside what_changed), no raw log/span search, and no way to configure the comparison window or baseline.