codeloop_verify
Run the verification suite (lint + tests + build + screenshot capture) on the current project. Returns a structured pass/fail report and a run_id you can pass to diagnose / gate_check.
When to use codeloop_verify
After every code change. The verify step runs lint + tests + build + (optional) screenshot capture and returns a run_id you pass to diagnose / gate_check.
Parameters
| Name | Type | Required | Description |
|---|---|---|---|
| scope | enum (full | affected) | no (default: full) | Run the full suite or only the checks affected by recent changes. |
| platform | enum (flutter | web | mobile | xcode | android | dotnet | auto) | no (default: auto) | Force a specific platform; defaults to auto-detect. |
| project_dir | string | no | Absolute path to the project root. |
What you get back
{ run_id, pass_count, fail_count, warning_count, artifact_paths, summary, suggestion }
Typical prompts that trigger it
Cursor + Claude Code call codeloop_verify automatically when your prompt looks like one of these:
- “Run codeloop_verify after the change.”
- “Build the project and run the test suite.”
- “Verify nothing is broken before opening the PR.”
FAQ
What does codeloop_verify do?
Run the verification suite (lint + tests + build + screenshot capture) on the current project. Returns a structured pass/fail report and a run_id you can pass to diagnose / gate_check.
When should an AI agent call codeloop_verify?
After every code change. The verify step runs lint + tests + build + (optional) screenshot capture and returns a run_id you pass to diagnose / gate_check.
What does codeloop_verify return?
{ run_id, pass_count, fail_count, warning_count, artifact_paths, summary, suggestion }
Is codeloop_verify part of the CodeLoop MCP server?
Yes. CodeLoop ships as an MCP server installed via `npx codeloop init`. Cursor and Claude Code call it automatically once the User Rule (Cursor) or CLAUDE.md (Claude Code) is in place.