Pant Antagonist Router

PANT Antagonist Router

Purpose

Choose the smallest useful PANT scenario, mode, entry path, pressure level, and optional companion. Route by the decision or preparation job—not by stakeholder title, framework name, industry, or desired aggression.

When to Use

Use when the user describes a messy situation, asks which PANT skill fits, combines an artifact with a meeting, or lacks the vocabulary to distinguish strategy, roadmap, investment, market, plan, artifact, and rehearsal work.

When Not to Use

If the user explicitly invokes a fitting specialist, use it. Do not insert the router as ceremony. Do not recommend multiple skills merely because they are compatible. A companion must perform a distinct second job.

Supported Modes and Entry Paths

• Feedback: Route to artifact or argument review.
• Dialogue: Route to rehearsal.
• Direct: Infer from the supplied context and start or recommend immediately.
• Guided: Explain the distinction in plain language and ask one question only when the answer changes the route.
• Best effort: State assumptions, choose provisionally, and let the user correct the route without restarting.

Decision Being Prepared For

Prepare the routing decision: which single scenario best matches the product decision or conversation, and which interaction settings let the user begin with the least unnecessary effort.

Typical Audiences

Any product practitioner using PANT, especially first-time users, technical practitioners moving into product, and experts arriving with compact context.

Inputs

Best input is the user's natural-language situation plus any artifact, decision, audience, stakes, and desired kind of help. Minimum input is a sentence such as “I have a difficult roadmap meeting tomorrow.” Do not demand taxonomy knowledge.

Direct example:

Use pant-antagonist-router. I have a TAM slide and need the CFO to fund a pilot. Tell me which skill to start with and whether a second skill adds a distinct job.

Sparse example:

I am walking into a difficult review and do not know which PANT skill I need.

Quick Start

1. Read the bundled catalog and trigger boundaries.
2. Identify the primary object: decision artifact, investment, estimate, plan, or conversation.
3. Identify the immediate job: strengthen substance, prepare an executive ask, or rehearse the room.
4. Choose one primary skill and infer mode, entry path, and pressure.
5. Ask one clarifying question only if its answer can change that selection.

Review Workflow

1. Honor explicit fit. If the user names a specialist and the context fits, use it without rerouting.
2. Name the immediate decision job. Distinguish review of substance from rehearsal of interaction.
3. Select the specialist. Match the dominant decision model in the bundled catalog.
4. Prefer one skill. Do not compose merely because skills list one another in combine_with.
5. Use fallbacks deliberately. Select the artifact generalist only when no specialist materially improves review; select the meeting generalist only when no specialist rehearsal fits.
6. Infer settings. Feedback for artifact review, dialogue for live practice; direct for sufficient context, guided for teaching needs, best effort for important ambiguity.
7. Set pressure proportionally. Default constructive or skeptical; use resistant or hostile-room only when requested or strongly implied by the preparation need.
8. Add a companion only for a second job. Example: roadmap feedback first, executive-decision rehearsal second.
9. Explain briefly. State why the primary fits and why the nearest neighbor does not.
10. Begin when possible. If the skill is available, transition into it rather than ending with a menu.

Dialogue Protocol

Routing dialogue is not an intake interview. Ask at most one question at a time and only when the answer materially changes scenario, mode, or pressure. Once the route is clear, stop routing and begin the selected work. Honor user corrections immediately.

Output

Produce a compact Routing decision:

• Primary skill
• Why it fits
• Why the nearest neighbor does not
• Mode, entry path, and pressure
• Material assumptions
• Optional companion and distinct second job, if any
• One clarifying question only when still decision-changing
• Suggested invocation or immediate transition

Use template.md only when catalog assessment needs to be recorded or tested.

Failure Modes

• Asking the user to select lenses, frameworks, or categories before helping
• Routing by stakeholder title or artifact filename alone
• Recommending every plausible skill
• Treating combine_with as a hard dependency
• Choosing the executive skill for every meeting with an executive
• Choosing roadmap review for backlog or delivery planning
• Choosing a fallback when a specialist clearly fits
• Continuing routing after the route is clear
• Claiming certainty when supplied context leaves a real boundary ambiguous

Evaluation Requirements

Evaluate every specialist and fallback trigger; neighboring negative cases; ambiguous mixed jobs; explicit invocation; direct, guided, and best-effort behavior; mode and pressure selection; minimal questioning; single-skill preference; optional two-stage composition; no hard dependencies; and transition into useful work.

Supporting Files

• Routing decision template
• Catalog and trigger boundaries
• Sources and adaptations
• Bundled interaction protocol
• Worked example
• Weak example

Provenance

Derived from the repository's scenario-first architecture, router rules, canonical metadata, and trigger-boundary evals. The bundled catalog is a standalone routing reference, not an installation dependency on every skill it names.