FaultDebug
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| --root | Yes | Absolute path to the artifact directory. The MCP server restricts artifact access to this root. |
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 |
|---|---|
| get_capabilitiesC | Return the bounded v1.0 capability and evidence contract. |
| check_storeB | Inspect store health and migration compatibility without writing. |
| open_faultD | – |
| get_thread_traceD | – |
| get_rpc_traceC | Return semantic RPC observations and unresolved evidence limits. |
| collect_faultsD | – |
| get_timelineD | – |
| get_process_relationsC | Join only explicit IPC/RPC endpoint evidence across artifacts. |
| get_evidence_summaryC | Summarize observed, static-candidate, and unresolved evidence. |
| get_incident_reportC | Return a read-only incident report with separate evidence classes. |
| get_session_manifestC | Return a read-only session manifest with separate evidence classes. |
| get_process_participantsC | Return read-only process participant metadata for a session. |
| list_sessionsC | List deterministic session summaries from a read-only evidence store. |
| get_sessionC | Return one session projection without merging evidence classes. |
| get_provenanceC | Return provenance and trust rows for local evidence artifacts. |
| list_incidentsC | List bounded agent-declared incidents without causal inference. |
| get_incident_sliceC | Return bounded process/artifact/event/source evidence for one incident. |
| get_unresolvedC | Return only unresolved evidence for an agent-declared incident. |
| get_source_evidenceC | Return exact file/line/hash citations from a verified source bundle. |
| discover_artifactsD | Discover and validate bounded artifact spool entries. |
| query_artifact_indexD | – |
| resolve_addressesD | – |
| get_function_sourceD | – |
| get_call_relationsD | – |
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 24 tools
Many tools share overlapping read-only reporting purposes (e.g., get_incident_report, get_session_manifest, get_incident_slice, get_evidence_summary) and eight tools lack descriptions entirely, making it hard to distinguish trace-related tools like get_thread_trace, get_rpc_trace, get_timeline, and get_call_relations. Boundaries are unclear for a significant portion of the set.
All tool names use snake_case with a consistent verb_noun structure (get_, list_, check_, open_, collect_, discover_, query_, resolve_). No mixing of conventions or casing styles.
With 24 tools, the set is on the heavy side for a fault-debugging server, and several tools appear to overlap in function, suggesting opportunities for consolidation. Still, the domain is complex enough that many capabilities are plausibly needed.
The surface covers sessions, incidents, traces, artifacts, source citations, and provenance, and includes tools for opening/collecting faults, which suggests broad lifecycle coverage. Some write/update/delete operations may be missing, but the read-focused debugging scope appears mostly complete.