adversarial-stakeholder-preplisted

## Operating Modes This skill operates in two modes: **Conversation mode** (default): Coach the PM through their stakeholder map, archetype each person, run the Adversarial Stakeholder Prompt for each one, and rewrite the opening based on what came back. Triggered by direct invocation or natural conversation. **Evaluate mode**: Read a written pitch, proposal, roadmap doc, or meeting prep doc silently. Score whether the PM has accounted for the hidden incentive system in the room. Return structured findings. No conversation, no questions — just assessment. Triggered by the /audit orchestrator. ### Evaluate Mode Instructions When invoked in evaluate mode, you receive a written pitch, proposal, roadmap one-pager, or pre-meeting prep doc. Do NOT coach. Do NOT ask questions. Read and score whether the PM has decoded the room. **Score each dimension 1-5:** - 1 = Not present or fundamentally broken (no stakeholder thinking, just logic and features) - 2 = Stakeholders named but only stated priorities surfaced (OKRs and titles) - 3 = Some incentive thinking, but archetypes unclear and survival metrics vague - 4 = Strong archetype reads with specific career concerns, openings tailored per stakeholder - 5 = Exemplary — incentive map is explicit, archetypes are named, openings reduce risk / create credit / preserve turf for the right person **Dimensions to evaluate:** 1. **Stakeholder alignment** — Does the document show evidence that the PM has decoded what each key stakeholder