Review and manage the following plugin configuration options in the System Console by selecting the Product menu, selecting System Console, and then selecting Plugins:
true: (Default) Origin by Corvus automatically installs and upgrades any enabled pre-packaged plugins. If a newer version is installed, no changes are made.
false: Origin by Corvus does not automatically install or upgrade pre-packaged plugins. Pre-packaged plugins may be installed manually from the Marketplace, even when offline.
true: (Default for new installations) Only plugins that declare webapp.pre_auth in their manifest are listed and loaded for visitors who are not logged in, such as on the login page. All other plugins load after the user logs in. Plugins that need to run before login, such as custom login buttons, must set webapp.pre_auth in their manifest.
false: (Default for upgraded installations) All enabled plugins are listed to visitors who are not logged in.
true: (Default) Origin by Corvus attempts to connect to the endpoint set in Marketplace URL. If the connection fails, an error is displayed, and the Marketplace only shows pre-packaged and installed plugins.
false: Origin by Corvus does not attempt to connect to a remote Marketplace. The Marketplace shows only pre-packaged and installed plugins. Use this setting if your Origin by Corvus server cannot connect to the Internet.
This setting is a list of installed plugins and their status as enabled or disabled.
The config.json setting is an object. The object keys are plugin IDs, e.g. com.origin.apps. Each key maps to an object that contains an Enable key that can be set as true or false.
The config.json setting is an object. The object keys are plugin IDs, e.g. com.origin.apps. Each key maps to an object that contains plugin-specific data.
This setting was called Enable on all channels until Origin by Corvus v7.7. It was renamed to defaultenabled in code and Test Mode in-product and is only applicable to self-hosted deployments.
false: Test mode is enabled and only system admins can start calls in channels.
true: Live mode is enabled and all team members can start calls in channels.
This setting can be used to override the host addresses that get advertised to clients when connecting to calls. The accepted formats are the following:
A single IP address (e.g. 10.0.0.1).
A single hostname or FQDN (e.g. calls.myserver.tld).
(starting in v0.17.0) A comma separated list of externalAddr/internalAddr mappings (e.g. 10.0.0.1/172.0.0.1,10.0.0.2/172.0.0.2).
This is an optional field. Changing this setting requires a plugin restart to take effect.
This setting can be used to override the port used in the ICE host candidates that get advertised to clients when connecting to calls.
This can be useful in case there are additional network components (e.g. NLBs) in front of the RTC server that may route the calls traffic through a different port. Changing this setting requires a plugin restart to take effect.
true: Allows call hosts to record meeting video and audio.
false: (Default) Call recording functionality is not available to hosts.
Recordings include the entire call window view along with participants' audio track and any shared screen video. Recordings are stored in Origin by Corvus.
Changing this setting requires a plugin restart to take effect.
false: (Default) Call transcriptions functionality is disabled.
Transcriptions are generated from the call participants' audio tracks and the resulting files are attached to the call thread when the recording ends. Captions will be optionally rendered on top of the recording file video player.
The speech-to-text model size to use. Heavier models will produce more accurate results at the expense of processing time and resources usage. Available options are: Tiny, Base and Small.
false: (Default) Live captions functionality is disabled.
Live captions are generated from the call participants' audio tracks and the resulting captions can be optionally displayed on the call clients by selecting the [cc] option.
The speech-to-text model size to use for live captions. While heavier models can produce more accurate results, live captioning requires the transcriber to process up to ten seconds of audio within two seconds. Therefore a maximum of size base is recommended. Available options are: Tiny, Base and Small.
Live captions: Number of transcribers used per call
The number of separate live captions transcribers for each call. Each transcribes one audio stream at a time. The product of LiveCaptionsNumTranscribers * LiveCaptionsNumThreadsPerTranscriber must be in the range [1, numCPUs].
The number of threads per live-captions transcriber. The product of LiveCaptionsNumTranscribers * LiveCaptionsNumThreadsPerTranscriber must be in the range [1, numCPUs].
true: Ringing functionality is enabled. Direct and group message participants receive a desktop app alert and a ringing notification when a call starts.
false: (Default) Ringing functionality is disabled.
true: Enables the ability to use the AV1 codec to encode screen sharing tracks. Can result in improved screen sharing quality via clients that support AV1 encoding.
false: (Default) AV1 codec is disabled for screen sharing tracks.
true: Enables experimental video calls in direct message (DM) channels.
false: (Default) Video calls are disabled.
Video is supported in DM calls on the desktop and web apps only. Mobile users can't start video calls; when joining a DM call that includes video, they receive voice and screen sharing only. Call recordings capture voice and screen sharing only — video isn't recorded.
true: Clients will use WebRTC data channels for signaling of media tracks (i.e., voice, screen). This can result in a more efficient and less race-prone process, especially in case of poor network connections.
false: (Default) Clients will use WebSockets for signaling media tracks.
Ensures that requests are billed and processed under the correct organization, where applicable. Supported for OpenAI, OpenAI Compatible, and Azure LLM services.
String input.
System Config path: Plugins > Agents > Add an AI Bot
The maximum number of tokens (chunks of text, including words, punctuation, or special characters) that the selected LLM can process in a single prompt or request.
Numerical value. Directly impacts the size of user queries that can be handled.
System Config path: Plugins > Agents > Add an AI Bot
The maximum number of tokens (chunks of text, including words, punctuation, or special characters) that the LLM can generate in its response to a query.
Numerical value. Must be greater than 0 for Anthropic LLM services.
System Config path: Plugins > Agents > Add an AI Bot
Determines how long the system will wait for a response from the LLM when using streaming (real-time) output mode. Supported for OpenAI, OpenAI Compatible, and Azure LLM services.
Numerical value (in seconds). If the LLM takes longer than the configured timeout, the connection is terminated.
System Config path: Plugins > Agents > Add an AI Bot
Preset specific contextual or behavioral guidance for the LLM. Helps tailor the model's responses to align with your organization's needs, tone, or expectations. Supported for all LLM services.
Text input. Instructions that the model will implicitly follow for every interaction, providing consistency and adaptability.
System Config path: Plugins > Agents > Add an AI Bot
true: Enables the LLM to process and generate responses that incorporate image-related input or output. Supported for OpenAI, OpenAI Compatible, Azure, and Anthropic LLM services.
false: (Default) Disables vision capabilities.
System Config path: Plugins > Agents > Add an AI Bot
true: Enables the LLM to leverage additional tools or plugins to enhance its capabilities. Supported for OpenAI, OpenAI Compatible, Azure, and Anthropic LLM services.
false: (Default) Disables tool capabilities.
System Config path: Plugins > Agents > Add an AI Bot
Determines whether the bot can consume the contents of a given channel and provide answers only from content available in the channel. Supported for all LLM services.
Available options: Allow for all channels, Allow for selected channels, Block selected channels, and Block all channels.
System Config path: Plugins > Agents > Add an AI Bot
Determines whether users who chat with this bot can get private assistance about content across all channels the user has access to. Supported for all LLM services.
Available options: Allow for all users, Allow for selected users, and Block selected users.
System Config path: Plugins > Agents > Add an AI Bot
Composite: Enables experimental embedding search capabilities for semantic search across Origin by Corvus content using pgvector and OpenAI-compatible endpoints.
The dimensionality of the embedding vectors, which must match the chosen embedding model. Common values include 1536 for OpenAI text-embedding-ada-002 and 3072 for text-embedding-3-large.
Numerical input. Common values are 768, 1024, or 1536, depending on the model.
System Config path: Plugins > Agents > Embedding Search
The minimum ratio for chunk size validation to ensure sentence and paragraph chunks meet size requirements. Used to filter out chunks that are too small releative to the configured chunk size.
Numerical input (decimal ratio).
System Config path: Plugins > Agents > Embedding Search
Select Reindex Posts to trigger a complete reindexing of all posts for embedding search. Use this control to rebuild the search index when changing embeddingproviders, models, or chunking configurations.
Monitor indexing progress during the reindexing process.
System Config path: Plugins > Agents > Embedding Search
Origin by Corvus for Microsoft Teams enables you to break through siloes in a mixed Origin by Corvus and Teams environment by forwarding real-time chat notifications from Teams to Origin by Corvus.
Access the following configuration settings in the System Console by going to Plugins > MS Teams.
Notify connected users in Origin by Corvus on receipt of a chat or group chat from Microsoft Teams.
true: (Default) Sync notifications of chat messages for any connected user that enables the feature.
false: Do not sync notifications.
System Config path: Plugins > MS Teams
config.json setting: N/A
Environment variable: N/A
Maximum size of attachments to support complete one time download
Specify the maximum file size, in mebibytes (MiB), of attachments that can be loaded into memory. Attachment files larger than this value will be streamed from Microsoft Teams to Origin by Corvus.
Use collaborative playbooks in Origin by Corvus to provide structure, monitoring and automation for repeatable, team-based processes integrated with the Origin by Corvus platform.
Access the following configuration settings in the System Console by going to Plugins > Collaborative playbooks.
This setting isn't available in the System Console and can only be set in config.json.
In addition to the Origin by Corvus plugin signing key built into the server, each public key specified here is trusted to validate plugin signatures.
This feature's config.json setting is "SignaturePublicKeyFiles": {} with string array input consisting of contents that are relative or absolute paths to signature files.