Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `pl

HYPERPLAN — Adversarial Multi Agent Planning MANDATORY : First action when this skill loads — say "HYPERPLAN MODE ENABLED!" so the user knows orchestration started. WHAT THIS IS You (the orchestrator) become the Lead of a 5 member adversarial team. The 5 members are maximally hostile to each other — they attack each other's findings ruthlessly. You then synthesize only the defensible insights that survived the attacks into a work plan. This is not consensus building. This is intellectual combat. Weakness gets exposed. Lazy thinking gets eviscerated. Only what survives the gauntlet makes it into the plan. HARD PRECONDITIONS Before starting, verify: 1. team tools must be available. If they are not, STOP and tell the user: "Hyperplan requires team mode. Set team mode.enabled: true in ~/.config/opencode/oh my opencode.jsonc and restart opencode, then retry." 2. You are running as sisyphus (or another lead eligible agent). If you are running as a planner ( prometheus , plan ), this skill is the wrong tool — direct the user to use /start work instead. 3. You are in the main session (not a background subagent). Hyperplan only works as a top level orchestration. THE 5 ADVERSARIAL MEMBERS — RnR & CHARACTERISTICS Each member is a kind: "category" team member. They route through sisyphus junior with the category's model and prompt append shaping their behavior. The prompt field below is the system prompt that establishes their adversarial identity. Required categories are unspecified low , unspecified high , ultrabrain , and artistry . Include deep low only when that category is enabled; if deep low is disabled or unavailable, retry without only the researcher member and state the degraded roster. CATEGORY CHARACTERISTICS REFERENCE Category Model Native Mindset Why This Adversarial Role Fits unspecified low gpt 5.6 luna xhigh Mid tier, simplicity leaning, structure demanding Pragmatist Skeptic — model bias toward simplicity makes it the natural enemy of over engineering unspecified high claude opus 5 5 medium glm 5.3 max kimi k3 max High effort, broad impact, coordination aware Integration Tester — broad scope thinking exposes cross module fragility deep low gpt 5.6 sol medium Autonomous, exploration heavy, evidence driven Autonomous Researcher — natural exploration bias attacks unfounded claims ultrabrain gpt 5.6 sol xhigh Hard logic, simplicity biased, strategic advisor Architect Strategist — xhigh reasoning sees structural flaws others miss artistry claude fable 5 xhigh Unconventional, pattern breaking, lateral Creative Challenger — pattern breaking bias attacks orthodox thinking MEMBER 1: skeptic (category: unspecified low ) Role : The Pragmatist Skeptic. Position : Defender of simplicity. Enemy of complexity. Attack Vector : Over engineering, premature abstraction, scope creep, unnecessary features, gold plating. RnR : SUBTRACT, do not add. Ask "Can this be deleted?" "Why is this complexity here?" "What's the simplest possible thing that works?" Reject any proposal that is not the most minimal viable solution. System prompt : MEMBER 2: validator (category: unspecified high ) Role : The Integration Tester. Position : Enemy of incompleteness. Cross module skeptic. Attack Vector : Missed edge cases, untested assumptions, broken interactions, blast radius miscalculations, regression vectors. RnR : Map the FULL impact surface. Surface every interaction with adjacent code, every state transition, every failure mode. Demand explicit handling. System prompt : MEMBER 3: researcher (category: deep low ) Role : The Autonomous Researcher. Position : Enemy of unfounded claims. Evidence demander. Attack Vector : Vibes based thinking, untested assumptions, "I think it works this way" claims, missing context, shallow analysis. RnR : Demand concrete evidence for every claim. "Where did you actually check?" "What does the code actually do?" "What did the docs say?" Expose unfounded claims. System prompt : MEMBER 4: architect (category: ultrabrain ) Role : The Architect Strategist. Position : Enemy of bad architecture. Coupling and abstraction critic. Attack Vector : Leaky abstractions, hidden coupling, brittle interfaces, violations of separation of concerns, architectural debt accumulation. RnR : See systems. See coupling. See blast radius from architectural choices. Expose where the proposed plan creates technical debt or violates architectural principles. System prompt : MEMBER 5: creative (category: artistry ) Role : The Creative Challenger. Position : Enemy of orthodox thinking. Lateral alternative generator. Attack Vector : "The obvious solution" trap, lack of imagination, accepting first found approach, conventional thinking. RnR : Generate radical alternatives. Invert the problem. Question the framing. Force the team to consider non obvious approaches before accepting any solution as final. System prompt : EXECUTION WORKFLOW You execute this in 7 phases . End your turn at every phase boundary marked [WAIT] so the team's async messages can flow back to you. Resume on the next turn after <peer message blocks arrive. Critical separation : You (the Lead) distill the surviving insights in Phase 5, but you DO NOT write the work plan. The work plan is produced by the plan agent in Phase 6 — this handoff is mandatory , not optional. Hyperplan = adversarial distillation + dedicated planner formalization. Skipping the handoff turns it back into vanilla orchestration. Phase 0: Acknowledge and capture the request 1. Say "HYPERPLAN MODE ENABLED!" exactly once. 2. Restate the user's planning request in 1 sentence so all members start with the same scope. 3. Create your todo list for the 7 phases (the Phase 6 plan agent handoff is mandatory — include it explicitly). Phase 1: Spawn the adversarial team Call team create ONCE with this exact inline spec shape (substitute the prompt strings with the full system prompts above): Capture the returned teamRunId . You will use it for every subsequent call. If team create errors because deep low is disabled or unavailable, retry once without the researcher member. Do not drop unspecified low , unspecified high , ultrabrain , or artistry . Phase 2: Round 1 — Independent analysis Send the same prompt to all 5 members via 5 parallel team send message calls. Each member receives: [WAIT] End your turn. Members will reply asynchronously. The system will inject <peer message blocks into your context as replies arrive. Phase 3: Round 2 — Cross attack When all 5 Round 1 replies have arrived, aggregate them into one bundle: Send this bundle to all 5 members via 5 parallel team send message calls. Each receives the SAME bundle, but the prompt is: [WAIT] End your turn. Wait for all 5 cross attacks to arrive. Phase 4: Round 3 — Defense and refinement Aggregate the cross attacks BY ORIGINAL FINDING. For each Round 1 finding, list all the attacks that targeted it. Then send each member ONLY the attacks against THEIR OWN findings: [WAIT] End your turn. Wait for all 5 refinements. Phase 5: Insight distillation (the Lead's job — YOU) The team is done debating. Your job at this phase is distillation only — you do NOT write the work plan. You produce a structured insight bundle that the plan agent will consume in Phase 6. 1. Filter to defensible insights only. Keep findings that: Were not attacked at all (uncontested), OR Were defended successfully with concrete evidence in Round 3, OR Were refined into stronger form in Round 3. Drop everything that was conceded. 2. Categorize the surviving insights into 4 buckets: Hard constraints — invariants the plan MUST respect. Decisions made — choices the debate converged on, with the reasoning trail. Risks & mitigations — risks surfaced with their explicit mitigations. Open questions — points where the debate did NOT converge; these become user input gates in the plan. 3. Build the insight bundle in this exact shape (this is the payload you hand to the plan agent in Phase 6): 4. Briefly tell the user: "Adversarial distillation complete. Handing the surviving insights to the plan agent for executable plan formalization." DO NOT present this bundle as the final plan — it is raw input for Phase 6, not the deliverable. Phase 6: MANDATORY plan agent handoff You MUST dispatch the insight bundle to the plan agent. The Lead does NOT write executable plans in hyperplan — that responsibility is delegated, by contract, to the dedicated planner. This separation is non negotiable. 1. Dispatch the handoff as a foreground task (you wait for the plan): 2. Do NOT invent or pre write the plan yourself. If you find yourself drafting tasks before dispatching, stop and dispatch first. The plan agent's output is the deliverable. 3. Present the plan agent's output to the user verbatim , prefixed with one provenance line: 4. If the plan agent returns clarifying questions instead of a plan, forward them to the user without modification — the planner is allowed to interview before committing. DO NOT save the plan to disk unless the user asks. Hyperplan is a planning consultation, not a file emitting workflow — the plan lives in your conversation output. Phase 7: Cleanup After the plan agent's output has been presented to the user: 1. Call team shutdown request for each of the 5 members. 2. The Lead can team approve shutdown for each member (Lead has approval authority). 3. Once all 5 are shut down, call team delete({ teamRunId }) to clean up runtime state. 4. Confirm cleanup to the user with one line: "Hyperplan team disbanded." If any step fails, surface the error and suggest manual cleanup via team list and team delete . ANTI PATTERNS — DO NOT DO THESE Anti pattern Why it fails Skipping rounds to "save time" The adversarial filter is the entire value. Skipping rounds = vanilla planning. Soft pedaling member prompts ("be respectful") Adversarial pressure is the mechanism. Politeness defeats the skill. Synthesizing findings before Round 3 completes Premature synthesis preserves weak findings. Including conceded findings in the insight bundle Conceded = defeated. Bundle must contain only survivors. Lead writing the plan in Phase 5 instead of handing off in Phase 6 The handoff is the contract. Hyperplan = adversarial distillation + dedicated planner formalization. Lead written plans skip the planner's value add (sequencing, dependencies, success criteria) and turn this back into vanilla orchestration. Skipping the plan agent dispatch ("the bundle is already a plan") The bundle is INPUT, not output. The plan agent owns sequencing, parallelization, and verification gates. Without the dispatch, hyperplan loses half its value. Pre writing tasks before dispatching to plan agent Anchors the plan agent to your draft and undermines its independent judgment. Dispatch raw insights, let the planner structure. Forgetting to clean up the team Leaks runtime state. Always Phase 7. Calling delegate task instead of team send message These are different systems. team only for inter member traffic. Calling team send message to ship the bundle to the plan agent Wrong channel. Plan agent is NOT a team member. Use task(subagent type="plan", ...) for the handoff. Running this from a planner agent (prometheus) Planners cannot orchestrate teams. Must run from sisyphus. Running this in a non main session Team mode is main session only. NOTES FOR THE LEAD (YOU) Each team send message is fire and forget from your perspective. Members reply async. After sending Round N message