This skill should be used when the user asks to "/harness-status", "harness status", "where is the harness", "what round are we on", "harness progress", or wants to inspect the current state of a harness run. Reads .harness/state.json and presents a human-readable progress report.
Scanned 5/27/2026
Install via CLI
openskills install uppifyagency/claude-harness---
name: harness-status
description: This skill should be used when the user asks to "/harness-status", "harness status", "where is the harness", "what round are we on", "harness progress", or wants to inspect the current state of a harness run. Reads .harness/state.json and presents a human-readable progress report.
allowed-tools: [Read, Glob, Grep]
---
# Harness Status — State Inspector
## Overview
Inspect the current state of a harness run. Reads `.harness/state.json` and related files to present a clear progress report.
## Execution
### Step 1: Check for Harness Workspace
Look for `.harness/` directory in the current project. If not found, inform the user that no harness run exists in this project.
### Step 2: Read State
Read `.harness/state.json` and extract:
- Current phase (planning / building / evaluating / complete)
- Current round number
- Max rounds configured
- Active components (planner, evaluator)
- Brief (what was requested)
### Step 3: Gather Artifacts
Scan `.harness/` for all artifact files:
- `spec.md` — exists? When created?
- `build-summary-*.md` — how many rounds completed?
- `eval-report-*.md` — how many evaluations done? Latest verdict?
- `config.json` — any disabled components?
- `simplification-log.md` — any simplification tests run?
### Step 4: Present Report
```
Harness Status
==============
Brief: [user's original brief]
Phase: [current phase]
Round: [N] / [maxRounds]
Components:
Planner: [enabled/disabled]
Evaluator: [enabled/disabled]
Artifacts:
spec.md [exists/missing]
build-summary-1.md [exists]
build-summary-2.md [exists]
eval-report-1.md [ITERATE]
eval-report-2.md [PASS]
Latest Eval: [verdict summary or "No evaluations yet"]
Next Action: [What should happen next based on state]
```
### Step 5: Suggest Next Action
Based on state, suggest what to do:
- `planning` + no spec → "Run `/harness` to start planning"
- `building` + no build summary for current round → "Generator is in progress or interrupted. Resume with `/harness resume`"
- `evaluating` + no eval report → "Evaluator is in progress or interrupted"
- `evaluating` + ITERATE verdict → "Run next build round or accept current state"
- `complete` → "Harness run is finished. Review results in `.harness/`"
No comments yet. Be the first to comment!