Skip to content
Back to skills

Stakeholder Objection Simulation

ASecurity

Simulate stakeholder objections through role-play and political feasibility analysis to test whether the decision survives real-world opposition.

  • 417 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added May 27, 2026
researchtesting

Security analysis

A100/100

Scanned May 27, 2026

npx -y skills add yogsoth-ai/de-anthropocentric-research-engine --skill stakeholder-objection-simulation --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Stakeholder Objection Simulation?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Stakeholder Objection Simulation
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/yogsoth-ai-stakeholder-objection-simulation/badge)](https://www.skillsdirectory.com/skills/yogsoth-ai-stakeholder-objection-simulation)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: stakeholder-objection-simulation
description: Simulate stakeholder objections through role-play and political feasibility analysis to test whether the decision survives real-world opposition.
used-by: steel-manning
---

# Stakeholder Objection Simulation

**Purpose:** Anticipate and simulate the strongest objections that real stakeholders would raise against the convergence decision, testing political feasibility and identifying resistance vectors before implementation.

**When to use:**
- When the decision affects multiple stakeholder groups
- When implementation requires buy-in from parties not in the room
- When political dynamics could block execution
- When prior decisions failed due to unanticipated opposition

## Budget

| Metric | Minimum |
|--------|---------|
| Stakeholder perspectives | >= 3 distinct groups |
| Objection depth | Full argument construction per stakeholder |
| Feasibility assessment | Explicit for each objection |

## State Ledger

```yaml
stakeholders_simulated: []
objections_raised: {}
severity_ratings: {}
mitigation_strategies: {}
political_feasibility: null  # HIGH | MEDIUM | LOW
```

## Available Tactics

| Tactic | When to Deploy |
|--------|---------------|
| multi-perspective-attack | Default — assign and attack from stakeholder perspectives |
| adversarial-debate-protocol | When a specific stakeholder objection needs formal resolution |

## Available SOPs

- perspective-assignment — define stakeholder perspectives and interests
- perspective-attack — simulate attacks from each stakeholder
- steel-manning-synthesis — synthesize across all stakeholder objections
- advocate-construction — build case for stakeholder's preferred alternative
- judge-verdict — assess whether objections are fatal

## Execution Guidance

1. Identify all affected stakeholder groups (minimum 3)
2. Deploy multi-perspective-attack with stakeholder-specific briefs
3. For each HIGH severity objection, assess mitigation feasibility
4. If any objection is both HIGH severity and LOW mitigability, escalate
5. Produce political feasibility rating with evidence

## Output Format

```yaml
strategy: stakeholder-objection-simulation
stakeholders_simulated: <count>
objections:
  - stakeholder: <group>
    objection: <core argument>
    severity: HIGH | MEDIUM | LOW
    mitigable: true | false
    mitigation: <if applicable>
political_feasibility: HIGH | MEDIUM | LOW
blocking_objections: []
recommended_accommodations: []
```

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.