generated: '2026-07-21'
method: searched
source: https://docs.vijil.ai/developer-guide/cli/setup
name: vijil
binary: vijil
package: vijil-console
install:
  - pip install vijil-console
  - pipx install vijil-console
config: ~/.vijil/config.yaml
docs: https://docs.vijil.ai/developer-guide/cli/setup
description: >-
  First-party CLI for the Vijil AI trust platform. Points at a Vijil Console
  gateway, authenticates via JWT, and drives the full agent-trust lifecycle
  (register agents, evaluate, protect with Dome, red-team, telemetry). Also the
  runtime behind the vijil-mcp MCP server.
command_groups:
  - group: auth
    docs: https://docs.vijil.ai/developer-guide/cli/setup
    commands:
      - {name: 'auth init', description: Configure the Console gateway URL}
      - {name: 'auth login', description: Log in and store JWT credentials}
      - {name: 'auth logout', description: Clear stored credentials}
      - {name: 'auth change-password', description: Change your account password}
  - group: team
    commands:
      - {name: 'team list', description: List teams you belong to}
      - {name: 'team use', description: Set the active team context}
  - group: agent
    docs: https://docs.vijil.ai/developer-guide/cli/agents
    commands:
      - {name: 'agent create', description: Create an Agent configuration}
      - {name: 'agent list', description: List Agent configurations}
  - group: harness
    docs: https://docs.vijil.ai/developer-guide/cli/evaluate
    commands:
      - {name: 'harness list', description: List available standard Harnesses}
  - group: eval
    docs: https://docs.vijil.ai/developer-guide/cli/evaluate
    commands:
      - {name: 'eval run', description: Start a trust Evaluation (optionally --wait)}
      - {name: 'eval results-detail', description: Retrieve Trust Score and per-Harness results}
      - {name: 'eval report', description: Generate a Trust Report}
  - group: protect
    docs: https://docs.vijil.ai/developer-guide/cli/protect
    commands:
      - {name: 'protect', description: Configure Dome Guardrail configs and run detections}
  - group: red-team
    docs: https://docs.vijil.ai/developer-guide/cli/red-team
    commands:
      - {name: 'red-team', description: Manage Personas, Policies, and run adversarial campaigns}
  - group: telemetry
    docs: https://docs.vijil.ai/developer-guide/cli/telemetry
    commands:
      - {name: 'telemetry', description: Query the Trust Dashboard and observability telemetry}
key_flows:
  - name: Evaluate an agent end to end
    steps:
      - vijil auth init --url https://<console-gateway>
      - vijil auth login
      - vijil team use <team_id>
      - vijil agent create --agent-name "My Agent" --model-name gpt-4o --agent-url <url> --api-key <key>
      - vijil eval run --agent-id <id> --harness-names '["safety","security"]' --sample-size 50 --wait
      - vijil eval results-detail <eval_id> --json
      - vijil eval report <eval_id>