Skip to main content
Day to day, you don’t need this page: your agent runs these commands through the skill. It exists for the moments a human drives Vise directly — reviewing recorded evidence, gating a release in CI, or peeking under the hood. For the complete list and every flag, run vise --help or vise <command> --help. Most agent-facing commands have an MCP tool equivalent in snake_case; trust signing and Managed Placement installation are intentionally CLI-only.
Most commands take a project path (default: the current directory) and read or write state under sp-vise/. The ones reviewers reach for most: vise impact, vise status, vise check --ci, and vise passport.

Explore and plan

For broad requests that span several surfaces (feed + chat + profile), Vise sequences them:

Day-2 and release assurance

See Review and release an integration for the recommended handoff flow.

Validate and verify

See Read the result for what each state means.

Evidence

Engagement history

Project sensors

Docs and SDK facts

Experience planning and outcome evidence (advisory)

These commands do not upload customer data, assign a calibrated score, change recommendations automatically, or affect vise check exit codes.

Design (advisory)

Skill and MCP

How Vise works

The loop behind these commands and how to read results.

Run Vise as an MCP server

The MCP tool equivalents of these commands.