MCP Clients / Telara CLI
Telara CLI
A command-line tool that connects your agentic coding tools to Telara in seconds. Log in, and the CLI automatically configures Claude Code, Cursor, and Windsurf with your organization's knowledge and integrations.
Overview
Why use the CLI?
Instead of manually copying API keys and editing JSON config files for each tool, the CLI handles everything: authentication, configuration selection, and writing the correct config files for your agent tools.
Run telara login and your installed tools are automatically connected to your organization’s knowledge base.
Organization defaults from your admin, your personal preferences, and per-project overrides — all resolved automatically.
API keys are generated and stored internally. You never see or handle raw keys — they’re written directly into tool configs.
Install
Installation
Install using your platform's package manager. The CLI is a single binary with no runtime dependencies.
Authentication
Logging in
The CLI authenticates against your Telara account. It opens your browser for sign-in (using your existing SSO if configured), then automatically connects your installed tools.
On first login, the CLI automatically detects your installed coding tools (Claude Code, Cursor, Windsurf) and connects them to your organization's Telara knowledge base. No manual setup required — your admin's data sources and permissions apply immediately.
On macOS, your session is stored in Keychain. On Linux, it uses your system's secret store. API keys are generated and managed internally — you never need to copy or paste raw keys.
Session Management
Logging out and back in
When you log out, your tool connections are saved privately. When you log back in, they're restored automatically — no extra steps.
Each account's saved connections are stored separately. Configurations belonging to a different account are never applied. If you log in under a different organization, organization-managed connections are skipped until you log back into the original account.
Configuration Layers
Three layers of configuration
Telara configs are applied in three layers. Higher layers override lower ones. This lets admins set organization defaults while individuals and teams customize as needed.
Set by your organization admin. Applied automatically when you log in. You can't remove it, but you can override it with Layer 2 or 3.
Scope: Organization-wideYour personal configuration choice. Set with telara config global — applies to all your projects. Overrides the managed config.
Scope: All your projectsPer-directory config that activates when your tools open a specific project. Overrides both managed and global configs. Set with telara config project.
Scope: Single project directoryStatus
See what's configured
Run telara configwith no arguments to see what's active at each layer and which tools are connected.
Per-Project
Project-level configuration
Use telara config project inside a project directory to set a configuration that activates only when your agent tools open that project. Ideal when different repos need different data sources.
The generated config files contain API keys that are managed automatically by the CLI. Add them to your .gitignore to prevent accidentally committing credentials to source control.
Reference
Commands
Beyond Local
Claude Web, CI/CD, and headless agents
The CLI is for local development machines. For environments where no CLI is available, Telara connects directly through the platform.
Your admin adds Telara as an Organization Connector in the Anthropic Admin Console. Every team member sees Telara tools in the Claude sidebar automatically — no setup per person.
Set the TELARA_API_KEY environment variable in your pipeline. Any agent framework that supports MCP will connect to Telara using the key.
Whether the connection comes from your laptop, a browser, or a CI server, the same permission policies, approval flows, and audit logging apply.
Troubleshooting
Diagnosing issues
Run telara doctor to check your authentication, connectivity, and tool configuration in one step.
Finding integration actions
The Telara CLI sees a fixed set of Telara tools when it connects — knowledge search, graph traversal, archive reads, task tracking, and two discovery tools. Your connected platforms are not all listed up front. GitLab alone contributes 107 actions and Jira 84; loading every schema into the model on each request would crowd out the context you actually want it thinking with.
Describe the task in plain language — "comment on a Jira issue", "open a merge request". Matching actions become callable by name straight away. Only actions your organization's policy permits are returned.
Returns one action's full parameter schema. Worth calling before invoking an action whose arguments are not obvious.
If you already know the action you want, telara_execute_action runs any permitted action directly — pass integration, action and params, with no search step. You can also browse the telara://integrations/available resource. Policy is enforced server-side on every call, whichever route you take.