harness-experiment
100 / 100 · Low risk
Fine to try.
Run a harness change as a controlled experiment, with a small end-to-end eval, one change at a time, repeated runs, and looking at flipped tasks, request count, tokens and cache hits, not just the average score. Use before and after changing prompts, tools, skills, subagent layout, compaction or other agent scaffolding, or when asked whether a change helped.
What this skill can do
We found nothing in the files we read that matches our rules. That does not prove the skill is safe: see the limits of static scanning.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'harness-experiment' ] || [ -L ~/.agents/skills/'harness-experiment' ]; then echo 'A skill folder named harness-experiment already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/MiniMax-AI/mcode-stack' "$d" && git -C "$d" checkout --quiet '0db1b4fb3b6270e173080d87b72c28516b299ddd' -- 'stacks/ronny/skills/harness-experiment' && mkdir -p ~/.agents/skills && cp -R "$d"/'stacks/ronny/skills/harness-experiment' ~/.agents/skills/'harness-experiment' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'harness-experiment' ~/.claude/skills/'harness-experiment' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
[](https://zalavijaysinh.github.io/skilltrust/repos/MiniMax-AI/mcode-stack/)
Score history
| Date | Version | Score |
|---|---|---|
| 2026-11-12 | 0db1b4f | 100 |
| 2026-11-08 | 88f84be | 100 |
| 2026-11-07 | 88f84be | 100 |