runboard-assess

name	runboard-assess
description	Run a Runboard 9-dimension maturity self-assessment with a technical leader, then persist it via the runboard CLI. Use when someone wants to assess or score their engineering maturity.

Runboard: Assess

You conduct the assessment conversation. The CLI computes and stores everything — you never calculate scores, averages, or constraints yourself.

Steps

1. Ensure the workspace exists. If .runboard/ is absent, run:

   runboard init
   

2. Read the anchors from .runboard/rubric.yaml. For each of the nine dimensions (build.team, build.tools, build.techniques, run.team, run.tools, run.techniques, plan.team, plan.tools, plan.techniques):

   • Show the 1–5 behavioural anchors.
   • Ask what best matches today. Suggest a level, but let the leader decide.
   • Capture a trajectory: up, flat, down, or volatile.
   • Capture one line of evidence.

3. Persist the assessment in a single non-interactive call. Pass every dimension:

   runboard assess \
     --set build.team=2:flat:"All devs contracted; no retained capability" \
     --set build.tools=3:up:"CI introduced last quarter" \
     --set build.techniques=2:flat:"Cadence inconsistent" \
     --set run.team=2:flat:"Informal on-call; knowledge concentrated" \
     --set run.tools=2:up:"Monitoring being extended" \
     --set run.techniques=3:flat:"Defined incident process" \
     --set plan.team=2:up:"CEO comms strengthening" \
     --set plan.tools=1:flat:"No roadmap tool or analytics" \
     --set plan.techniques=2:up:"Strategy being established"
   

4. Show the result:

   runboard board
   

   Read back the average and the named binding constraint from the CLI output. Do not restate numbers you computed yourself.

Rules

• All nine dimensions are required in one assess call.
• Levels are 1–5; trajectory is one of up/flat/down/volatile.
• If the CLI reports a validation error, fix the offending dimension and re-run.
• Never overwrite an existing same-day assessment without confirming --force with the user.