Entelligence connects to a Nagent workspace with no credentials. Once it is connected, agents can call 2 Entelligence actions: "Add a new repository" and "Interact with the repository agent". Nothing is enabled on connect: each action is allowed one at a time, and an action with side effects runs or waits for a person according to the agent's level.
Every operation an agent can call against Entelligence, with input parameters and output schema.
ENTELLIGENCE_ADD_A_NEW_REPOSITORYAdd a GitHub repository to Entelligence for code analysis and AI-powered insights. The repository will be analyzed asynchronously. After adding, you can use the 'interact_with_the_repository_agent' action to ask questions about the codebase. Returns a repo_uuid that can be used to track the repository.
Input parameters
The full GitHub repository URL (e.g., 'https://github.com/owner/repo'). Must be a valid, accessible GitHub repository URL.
A display name for the repository (e.g., 'my-react-app'). This can be different from the actual repository name.
The branch to analyze (e.g., 'main', 'master', 'develop'). If not specified, the default branch will be used.
Set to true for public repositories, false for private repositories. For private repos, you must also provide an AccessToken.
UUID of the user adding the repository. If not provided, the system will generate one.
Set to true to ingest and analyze pull requests from the repository. Defaults to false.
Set to true to ingest and analyze documentation files from the repository. Defaults to false.
GitHub personal access token for private repositories. Required when IsPublic is false.
URL of an existing vector database to use for this repository. If not provided, a new vector DB will be created.
Set to true to ingest and analyze issues from the repository. Defaults to false.
Set to true if this repository is a merged/combined repository from multiple sources. Defaults to false.
URL to external documentation for this repository (e.g., 'https://docs.example.com'). Useful for linking to official docs.
UUID of the organization this repository belongs to. Used for grouping repositories under an organization.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
ENTELLIGENCE_INTERACT_WITH_THE_REPOSITORY_AGENTAsk questions about a codebase using Entelligence's AI-powered Repository Agent. Requires a repository to be added first using the add_a_new_repository action. The agent can answer questions about code architecture, functionality, dependencies, and help with understanding the repository structure.
Input parameters
Conversation history as a list of message objects with 'role' ('user' or 'assistant') and 'content' keys. Use an empty list \[\] for the first question in a conversation
The question to ask the Repository Agent about the codebase, e.g., 'What is the main purpose of this repository?' or 'Explain the authentication flow'
Display name or identifier of the user. Optional, used for personalization
The repository UUID (repo_uuid) returned by the add_a_new_repository action. This identifies which repository to query
Enable advanced agent features for more sophisticated code analysis. Defaults to True. Set to False for faster, simpler responses
GitHub username of the person asking the question. Optional, used for personalization and access control
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
Agents can call 2 Entelligence actions on Nagent: "Add a new repository" and "Interact with the repository agent". Add a new repository: Add a GitHub repository to Entelligence for code analysis and AI-powered insights. Each action is listed on this page with its input parameters and its output.
Entelligence connects with no credentials, under your workspace's own connection. Nothing is enabled on connect: each action is allowed one at a time and can be scoped to the agents that need it.
Add a new repository takes 3 required inputs: Url, Name and IsPublic. It also takes 10 optional inputs: Branch, UserUUID, IngestPrs, IngestDocs, AccessToken, VectorDBURL, IngestIssues, IsMergedRepo, DocumentationUrl and OrganizationUUID. It returns data, error and successful.