Upstream Consistency Verification

Verifies downstream artifacts faithfully represent upstream pipeline source materials in YAML reports.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill upstream-consistency-verification-gozonerd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Upstream Consistency Verification
Source: https://github.com/gozonerd/shadow-ai-assessment/tree/main/.claude/skills/dmis/task_type/TT-09_Upstream_Consistency_Verification
Command: npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill upstream-consistency-verification-gozonerd

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Downstream deliverables like outlines and prose can silently distort, omit, or misrepresent upstream source materials such as domain syntheses and evidence bases. This Skill cross-checks downstream artifacts against upstream pipeline outputs to detect distortions, omissions, and fidelity losses before content reaches publication. ## Core Features & Use Cases - Upstream-to-Downstream Mapping Verification: Explicitly maps where each piece of upstream content appears in the downstream artifact and assesses coverage. - Distortion and Omission Detection: Identifies paraphrasing that changes meaning, flags silently omitted upstream content, and evaluates fidelity against defined standards. - Accuracy Rule Compliance: Enforces 9 Critical Accuracy Rules (ACC-001 through ACC-009) covering terminology, attribution, and methodology correctness, validated by a Python script. - Use Case: After generating a prose section from an outline and domain synthesis, run this verification to produce a YAML report confirming 94%+ of upstream content is represented with no meaning drift. ## Quick Start Verify that my generated prose section faithfully represents the upstream outline and domain synthesis, and produce a TT-09 consistency report in YAML.

Frequently Asked Questions about Upstream Consistency Verification

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

FAQPage Schema
How do I verify content consistency between pipeline stages?

Provide the upstream deliverables, the downstream artifact, and mapping documents, then run the verification methodology to produce a YAML report. The report maps each upstream content point to its downstream location and flags distortions or omissions.

What does an upstream consistency verification report contain?

The YAML report contains a consistency summary, upstream-to-downstream mapping verification, distortion detection, integration assessment, and fidelity analysis. It must conform to the TT-09 output schema and include accuracy_compliance entries for all nine rules.

How do I validate a TT-09 verification report?

Run scripts/validate_output.py with the report file path as argument. The script checks YAML validity, task_type, version, and confirms all nine accuracy rules ACC-001 through ACC-009 appear in the accuracy_compliance section.

What accuracy rules does the verification enforce?

Nine rules cover terminology and attribution accuracy, such as distinguishing CommCare from Dimagi, using DATS not DAD, EFD not RQ-RE, and correct convergence thresholds. All nine must be addressed in every verification report.

When should upstream consistency verification not be used?

It is designed for pipeline stages P3 through P5 where downstream content derives from upstream sources. It does not fit standalone content creation with no source materials, or factual verification against external real-world data.