silence-audit

Audit persona panels for under-represented groups and rate their material impact.

5|2|Updated Feb 15, 2026
One-click install
npx skills add https://github.com/NOMARJ/sigil --skill silence-audit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: silence-audit
Source: https://github.com/NOMARJ/sigil/tree/main/packs/discovery/skills/silence-audit
Command: npx skills add https://github.com/NOMARJ/sigil --skill silence-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Persona panels and internet-sourced evidence routinely over-represent vocal, English-speaking, tech-literate, and urban populations while under-representing groups whose absence can materially skew product decisions; this skill makes those omissions visible, assesses their impact, and prescribes mitigation paths so teams do not mistake majority internet signals for the whole population.

Core Features & Use Cases

  • Evidence inventory: Enumerates the types of sources present in a persona panel (reviews, forums, social media, research, analytics, interviews, etc.) and highlights missing source types.
  • Bias detection & impact assessment: Identifies commonly under-represented groups (non-English speakers, older adults, low-income users, people with disabilities, regulated-industry users, rural populations, minority groups) and rates the material impact as HIGH / MED / LOW.
  • Mitigation recommendations: Offers actionable options per missing group: label limitations, seek alternative evidence (census, NGO reports, academic studies), or flag as fieldwork priority with acceptance-criteria implications.
  • Use Case: During a DISCOVER run, run this audit after persona construction to prevent product features from being validated exclusively against a skewed evidence base.

Quick Start

Run a silence-audit on the current persona panel to identify missing groups, rate their material impact, and generate mitigation actions.

Frequently Asked Questions about silence-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find missing user groups in my persona research?▼

A silence audit detects under-represented groups in persona panels by inventorying present evidence sources and highlighting missing demographics. It rates the material impact of these omissions as HIGH, MED, or LOW to prevent skewed product decisions.

What is a silence audit in user research?▼

A silence audit in user research is a structured review that reveals missing voices in your evidence base. It enumerates present and absent source types, assesses the impact of under-represented groups, and recommends mitigation steps like seeking alternative data sources.

How do I assess bias impact in an evidence panel?▼

Assessing bias impact in an evidence panel involves checking for commonly under-represented groups such as older adults, low-income users, and rural populations. The audit rates the material impact of these missing groups to determine if majority signals mask the whole population.

When do I need to run a blind spot audit on personas?▼

Run a blind spot audit on personas during the DISCOVER workflow after persona construction. This prevents product features from being validated exclusively against a skewed evidence base that over-represents vocal, English-speaking, and tech-literate populations.

How do I mitigate missing evidence sources in persona panels?▼

Mitigate missing evidence sources in persona panels by labeling limitations, seeking alternative evidence like census or NGO reports, or flagging the gap as a fieldwork priority. The audit generates a structured report with actionable mitigation paths per missing group.

Does persona bias affect product decisions for non-tech users?▼

Persona bias materially affects product decisions when evidence over-represents tech-literate, urban populations while ignoring non-tech users. The audit quantifies this impact and outputs a blind spot rating to highlight risks of validating features against skewed internet signals.