Skip to Content
SettingsOverview

Settings

The Settings page controls workspace-wide preferences and integrations. Its tabs stay in the URL, so a link to a specific tab opens that tab directly.

Only people with a role in the workspace see it (settings.view, part of every role). Visitors of a public workspace, and members reading one without a grant, see neither the page nor the settings behind it; the gear on the workspace card and in the header is not shown to them. The AI providers and the tool servers the autopilot uses are settings in the same sense: listing them needs settings.view.

TabWhat it controls
GeneralThe workspace itself (name, address, description, categories, links, picture), software version, SQL limits, scan limits, language, timezone, time format, and the Hugging Face token
MCP ServersThe Classifyre MCP endpoint, bearer tokens, and client connection guides

General

General settings cover the workspace itself rather than AI behavior.

The workspace

Its name, URL slug, description, categories, links and picture — what the workspace directory shows on its card. The gear on the card opens this tab. Changing them needs workspace.settings; a new slug moves the page to the new address. Pausing and resuming are under Workers.

Software version

The version card shows the installed Classifyre release and whether an update is available.

SQL over tables

The limits of SQL over tables: rows per result, the time limit and rows per export. They apply to this workspace; who may query which table follows roles and data-reader grants.

Scan limits

Three limits govern how much scanning the installation does. Changing them needs workspace.settings and takes effect within seconds, with no restart.

SettingWhat it limitsHelm chart valueAll-in-one image variable
Scans running at onceScans this workspace runs at once; further scans wait in the queue. 0 means no workspace limit.api.env.MAX_CONCURRENT_RUNNERSMAX_CONCURRENT_RUNNERS
Automatic scans at onceScans in flight in this workspace before the adaptive scheduler asks for another one. 0 turns it off.api.env.AUTO_SCHEDULE_MAX_CONCURRENTAUTO_SCHEDULE_MAX_CONCURRENT
Runs kept per sourceFinished runs kept for each source of this workspace; older ones are removed. 0 keeps everything.api.env.MAX_RUNNERS_PER_SOURCEMAX_RUNNERS_PER_SOURCE

The limits are per workspace: each workspace saves its own values, and the settings of one never change another.

A value saved here replaces the one the Helm chart or the all-in-one image sets, for this workspace. Each field shows what the chart or image sets underneath it. Leave a field empty and save to follow the chart or image again.

Scans running at once is the exception to “replaces”: the chart or image value stays the ceiling for the whole installation, because every workspace shares the same machine. The workspace value caps the scans this workspace runs at once, so it can hold a workspace below the ceiling but never above it.

Language, timezone, and time format

These settings support Automatic browser detection or a fixed workspace-wide value.

SettingOptionsAutomatic behavior
LanguageAutomatic, English, GermanUses the browser language
TimezoneAutomatic or an IANA timezoneUses the browser timezone
Time formatAutomatic, 12-hour, 24-hourUses the browser hour-cycle preference

A user can override the displayed language from the header without changing the workspace default. See Instance Settings for the full resolution rules.

Hugging Face token

A Hugging Face token authenticates model downloads for audio transcription and Hugging Face custom detectors. It may come from the deployment or be saved on the General tab. Deployment-level credentials take precedence and cannot be replaced from the UI.

See Instance Settings for configuration and security details.


MCP Servers

The Classifyre Model Context Protocol (MCP) server exposes workspace data and operations to clients such as Claude, Cursor, and Codex CLI over a standard HTTP API. The tab controls the endpoint and its individual bearer tokens.

See MCP Server for client setup and troubleshooting.


AI configuration

AI providers are configured in Harness AI → Configuration, not on the Settings page. That is where you assign providers independently to the interactive Assistant and the autonomous Harness, create or edit provider credentials, and run provider tests.

See AI Providers for the complete setup guide.

Harness agents

When each autonomous agent runs — immediately on new data, batched, or only once the corpus has settled — is configured per agent in Harness → Agents, alongside its system prompt and tools. Instance-wide harness limits (run budgets, evidence thresholds, and how much context each run sees) are under Harness → Config.

See Harness Agents for what each mode and gate means, and for what to check when an agent is not running.

Last updated on