Hyperplan — Adversarial Multi-Agent Planning

Runs five mutually hostile AI agents through three rounds of cross-critique, then hands only the surviving insights to a dedicated planner agent for an executable plan.

AutomationAdvanced★ 69,329⑂ 5,715AI score 7/10Last updated: Sep 23, 2026

What it does

Hyperplan replaces consensus-building with intellectual combat.

Spawns 5 adversarial members: Pragmatist Skeptic (attacks over-engineering), Integration Tester (attacks missed edge cases and blast radius), Autonomous Researcher (demands file:line evidence), Architect Strategist (attacks hidden coupling and leaky abstractions), Creative Challenger (attacks orthodox framing)
Three debate rounds: ① independent findings → ② ruthless cross-attack → ③ defend / refine / concede
Distillation: conceded findings are dropped; survivors are bucketed into Hard Constraints, Decisions, Risks & Mitigations, and Open Questions
Mandatory handoff: the Lead never writes the plan — the insight bundle goes to the plan agent, which owns sequencing, dependencies, parallelization, and per-task success criteria
Cleanup phase: shutdown and delete the team run

Who it's for

opencode (oh-my-opencode) users who have team_mode enabled
Tech leads and senior engineers who want blind spots surfaced before writing any code
Teams planning expensive-to-reverse work like refactors or architecture migrations

Examples

"hyperplan: move the billing module to an event-driven design" — the skeptic deletes speculative abstractions, the integration tester enumerates breaking tests, the architect exposes hidden coupling, and the planner turns what survives into ordered tasks.
"adversarial plan for the new auth flow" — the creative member forces three alternatives onto the table while the researcher invalidates guesses about library behavior.
"hpp: plan this sprint's data migration" — unresolved contentions land in Open Questions and become explicit user-input gates in the final plan.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/HEAD/.agents/skills/hyperplan/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)

Download the ZIP with the button below.
In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.

↓ Download ZIP

Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the .agents/skills/hyperplan folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/hyperplan/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/code-yeongyu/oh-my-openagent.git /tmp/oh-my-openagent && mkdir -p ~/.claude/skills && cp -r /tmp/oh-my-openagent/.agents/skills/hyperplan ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

Open your terminal.
Clone the repo: git clone https://github.com/code-yeongyu/oh-my-openagent.git /tmp/oh-my-openagent
Create the skills directory: mkdir -p ~/.claude/skills
Copy the skill: cp -r /tmp/oh-my-openagent/.agents/skills/hyperplan ~/.claude/skills/
(opencode users) Set team_mode.enabled: true in ~/.config/opencode/oh-my-opencode.jsonc and restart opencode.
In a new main session, type hyperplan (or hpp) followed by what you want planned.
Note: without the team_* tools and a plan agent the skill will stop and tell you so; it only works from a top-level session, not a background subagent.

View source on GitHub ↗License: NOASSERTION