adversarial-consensus
100 / 100 · Low risk
Fine to try.
For a large or hard-to-reverse decision, or when the owner asks for it: several delegate_task children answer blind and independently, then each critiques the others' anonymised answers, and you judge — naming agreements, disagreements and what nobody checked. Never averages. Costs about 2N child runs.
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/'adversarial-consensus' ] || [ -L ~/.agents/skills/'adversarial-consensus' ]; then echo 'A skill folder named adversarial-consensus already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/theselfruleorg/polyrob' "$d" && git -C "$d" checkout --quiet '79043688317d0bdc7930758ea95f6dc1268ddf2c' -- 'data/prompts/skills/adversarial-consensus' && mkdir -p ~/.agents/skills && cp -R "$d"/'data/prompts/skills/adversarial-consensus' ~/.agents/skills/'adversarial-consensus' && 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/'adversarial-consensus' ~/.claude/skills/'adversarial-consensus' (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/theselfruleorg/polyrob/)
Score history
| Date | Version | Score |
|---|---|---|
| 2026-11-08 | 7904368 | 100 |