mcpField
W
watch-run

watch-run

github.com/coval-aiTesting & quality361 installsUpdated 1d ago
Repository

About watch-run

A skill for tracking Coval run progress and reporting completed simulation results. It guides users through selecting a run, watching its status, and summarizing outcomes.

Key features

  1. 01List recent runs when no ID is given
  2. 02Display live progress and status
  3. 03Summarize counts, duration, and results link

Use cases

  1. 01Monitor a selected Coval run
  2. 02Review simulation results after completion

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/coval-ai/coval-external-skills --skill watch-run

After installing, type/watch-runin Claude Code to confirm it loaded.

Details

From the project README

Watch Coval Run Progress

Monitor the progress of run $ARGUMENTS.

Workflow

Step 1: Identify Run

If no run ID provided, list recent runs:

coval runs list --page-size 10

Ask user to select a run to watch.

Step 2: Watch Progress
coval runs watch <run_id> --interval 2

This displays a live progress bar showing:

  • Completed / Total test cases
  • Current status (PENDING, IN QUEUE, IN PROGRESS, COMPLETED, FAILED)
  • Percentage complete
Step 3: Report Results

When the run completes, summarize:

coval runs get <run_id> --format json

Report:

  • Total simulations completed
  • Pass/fail counts (if available)
  • Duration
  • Link to results

Options

FlagDescriptionDefault
--intervalRefresh rate in seconds2

Statuses

StatusMeaning
PENDINGRun created, not yet queued
IN QUEUEWaiting for capacity
IN PROGRESSSimulations running
COMPLETEDAll simulations finished
FAILEDRun encountered an error
CANCELLEDUser cancelled the run

Comments

Markdown supported