Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
rootYesThe path to the approved source directory
providerYesThe provider type, e.g., 'local'
state_dirYesThe path to the local private state directory

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

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
team_drive_statusA

Check index coverage, current jobs, supported formats and configured limits. Counts describe the last inventory, not guaranteed current business totals.

team_drive_listA

Page through the indexed file inventory, including unsupported and failed files. Use all pages for a scope manifest; verify versions before evidence use.

team_drive_searchA

Find candidate files using names and indexed keyword content, including short Chinese terms and configured aliases. Search results are not authoritative or exhaustive business data; inspect/read before answering.

team_drive_inspectC

Verify the current source hash and page through indexed structure groups with segment offsets, format limitations and version. A recent modification time alone does not establish official status.

team_drive_readA

Read a bounded segment page with original locators and source version, verifying the current file. Optional selector: rows for CSV/TSV; sheet and rows for XLSX; page for PDF; slide for PPTX. start indexes selected segments. Continue with next_cursor and identical selector/expected_version; missing rows are not zero.

team_drive_exportA

Export all indexed segments of an explicitly versioned file to a controlled local JSONL artifact for deterministic processing. Check coverage and parser warnings first. Does not modify the source.

team_drive_refreshA

Queue a local inventory and incremental reindex job. Use force only to retry unchanged failures or deliberately rebuild the index. Source files remain read-only; poll job_status for completion.

team_drive_job_statusA

Read current or specified indexing job progress, coverage and failures. A completed scan can still contain failed or unsupported files.

team_drive_cancelA

Request cooperative cancellation of an indexing job. Already indexed files remain; resume by refreshing. Writes local job state only.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

A3.8/5.0

Scored across 9 tools

Disambiguation4/5

Each tool targets a distinct stage of the index lifecycle, and the descriptions actively draw boundaries (status = inventory totals/limits, job_status = live job progress; inspect = structure/hash verification, read = content segments). The only mild overlaps are status vs job_status and inspect vs read, both of which are clarified but still plausibly confusable at a glance.

Naming Consistency5/5

All nine tools use a uniform team_drive_<action> snake_case pattern with predictable verbs (list, search, inspect, read, export, refresh, cancel). No mixed conventions or abbreviations; an agent can guess new tool names reliably.

Tool Count5/5

Nine tools is well within the ideal range and each maps to a necessary distinct operation in the index-and-read workflow. Nothing feels redundant or bolted on.

Completeness5/5

The surface covers the full lifecycle: triggering/cancelling/refreshing indexing, monitoring job and overall status, discovering files via list and search, verifying via inspect, reading content pages, and exporting a versioned artifact. Sources are intentionally read-only, so the absence of write tools is correct rather than a gap.

Maintenance

ActivityMaintained
ResponsivenessNo issues