Presupposition Archaeology

Identifies and remediates ungrounded implicit assumptions in DATS narrative documents.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill presupposition-archaeology-gozonerd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Presupposition Archaeology
Source: https://github.com/gozonerd/shadow-ai-assessment/tree/main/.claude/skills/dmis/task_type/TT-16_Presupposition_Archaeology
Command: npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill presupposition-archaeology-gozonerd

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) components.

What problem does it solve? Documents often rely on unstated assumptions that the target audience may not share, creating comprehension gaps and undermining transparency. This Skill reads a narrative from the specified audience's perspective and surfaces every implicit assumption that is not grounded in the text, evidence, or definitions. ## Core Features & Use Cases - Assumption Extraction: Flags implicit factual, methodological, and definitional assumptions by document location, with grounding status (explicitly stated, implied only, external prerequisite). - Gap Classification & Remediation: Categorizes each gap as missing definition, citation, or justification, and drafts concrete language additions with priority levels. - Accuracy Rule Enforcement: Applies 9 critical accuracy rules (ACC-001 through ACC-009) covering CommCare, Dimagi, DATS, and EFD terminology, with YAML output validated by a Python script. - Use Case: Before distributing a DATS evaluation report to policy-level decision makers, run a presupposition audit to find where the text assumes knowledge of CommCare or EFD methodology and receive specific remediation language. ## Quick Start Audit the attached DATS narrative for implicit assumptions from the perspective of policy-level decision makers and produce a remediation report.

Frequently Asked Questions about Presupposition Archaeology

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

FAQPage Schema
How do I audit a document for unstated assumptions?

Provide the complete narrative text, an audience specification, and the evidence database. The Skill reads from the audience's perspective, extracts implicit assumptions by location, classifies their grounding status, and outputs a YAML audit with remediation language.

What does the presupposition audit output contain?

The YAML output contains an assumptions_identified list with location, grounding status, gap type, and audience impact, a grounding_assessment summary counting explicit, implied, and external assumptions, and a remediation section with recommended text additions and priorities.

How is the audit output validated?

Run scripts/validate_output.py with the output YAML file as argument. It checks required sections, validates grounding_status and priority enum values, and confirms grounding assessment fields are integers, exiting with an error list if invalid.

What are the accuracy rules ACC-001 through ACC-009?

They are terminology and factual constraints for DATS documents, such as distinguishing CommCare current use from history, Dimagi versus CommCare, DATS versus DAD, EFD versus RQ-RE, and convergence thresholds of alpha 0.70 or 0.75. All outputs must comply with them.

What are common mistakes when identifying assumptions?

Documented anti-patterns include auditing from an expert perspective instead of the target audience, flagging trivially obvious prerequisites, missing definitional or methodological gaps, and producing flags without actionable remediation guidance.