Idaho Journalism Corpus
Relevant source files
The following files were used as context for generating this wiki page:
The Idaho Journalism Corpus is the primary knowledge base and working record within the IDAHO-VAULT project. It functions as an extended memory system for Logan Finney, a reporter and producer at Idaho Reports / Idaho Public Television. The corpus is a collection of approximately 3,000 markdown notes in an Obsidian.md vault, version-controlled via GitHub, covering Idaho politics, government, and legislation.
Sources: Claude vault project memory 3-28-2026.md, VAULT-CONVENTIONS.from-pushable-2026-05-08.md:15-20
The corpus serves a "journalism operating system" where raw world inputs are transformed into publishable insights through a structured pipeline of AI-assisted intake, processing, and human verification. It provides an authoritative record of institutional accountability and state power in Idaho.
Sources: IDAHO-VAULT — END-TO-END JOURNALISM WORKFLOW (AI-ASSISTED, LOCAL-FIRST)v0.1.md:215-220, Claude vault project memory 3-28-2026.md
Corpus Architecture and Taxonomy
The corpus is organized into specific root folders that categorize the entities and sources relevant to Idaho reporting. This structure allows for a "census mode" analysis of interconnected repositories and topics.
Sources: LEVELSET-v2.md:52-65, CONTEXT-PASTE-BUNDLE.md:46-60
Core Directory Structure
| Path | Description |
|---|---|
GOVERNMENTS/ | Notes on Idaho legislative, executive, judicial, federal, and tribal entities. |
ORGANIZATIONS/ | Data on churches, companies, education, hospitals, and political parties. |
PEOPLE/ | Profiles and notes on public figures and elected officials. |
PLACES/ | Geographic locations including Idaho cities, counties, and geography. |
SOURCES/ | Raw material: news media, transcripts, hearings, and press releases. |
TOPICS/ | Subject-area notes such as fiscal health, elections, and education. |
Sources: VAULT-CONVENTIONS.from-pushable-2026-05-08.md:44-65, CONTEXT-PASTE-BUNDLE.md:48-60
Entity Relationship Mapping
The following diagram illustrates how raw sources are processed into specific entities within the corpus taxonomy.
Journalism Workflow
The corpus operates on a metadata-driven lifecycle where units of work (individual documents or notes) evolve through several stages of transformation.
Sources: IDAHO-VAULT — WORKFLOW.md:38-45
Data Lifecycle Stages
- Capture (Raw): External information (URLs, PDFs, transcripts) is brought into the vault, typically placed in
/INBOX/. - Structure (Processed): AI agents normalize formatting and identify key entities like bill numbers, sponsors, and fiscal impacts.
- Interpret (Analysis): The system connects related notes, identifies trends, and flags anomalies.
- Synthesize (Draft): Narrative-ready content is assembled into the
/DRAFTS/directory. - Verify (Human): Logan Finney reviews all claims against original sources to ensure accuracy before publication.
Sources: IDAHO-VAULT — END-TO-END JOURNALISM WORKFLOW (AI-ASSISTED, LOCAL-FIRST)v0.1.md:120-170, IDAHO-VAULT — WORKFLOW.md:80-140
Workflow Processing Sequence
The sequence diagram below shows the interaction between the Human Artificer, AI Agents, and the Vault Storage during document processing.
Metadata and Naming Standards
Reliable transformation within the corpus depends on strict naming conventions and YAML frontmatter to define the state of a note.
Sources: VAULT-CONVENTIONS.from-pushable-2026-05-08.md:85-90, IDAHO-VAULT — WORKFLOW.md:48-52
Naming Conventions
| Document Type | Format | Example |
|---|---|---|
| Legislative Bills | (YYYY) Bill Type Number.md | (2026) House Bill 24.md |
| News Articles | YYYY-MM-DD - Outlet - Title.md | 2024-01-15 - Idaho Statesman - Title.md |
| Hearings | YYYY-MM-DD - Committee.md | 2023-12-19 - GIAC meeting.md |
| People | Full Name.md | Brad Little.md |
Sources: VAULT-CONVENTIONS.from-pushable-2026-05-08.md:105-115, CONTEXT-PASTE-BUNDLE.md:75-85
Mandatory Metadata Fields
Every governed markdown note in the corpus must include these frontmatter keys to track its lifecycle and authority:
---
title: "Document Title"
updated: YYYY-MM-DD
status: draft | active | superseded | archived
authority: "Logan Finney"
tags: []
source: "URL or Origin"
---
Sources: VAULT-CONVENTIONS.from-pushable-2026-05-08.md:175-190, IDAHO-VAULT — WORKFLOW.md:55-65
Sourcing and Ethical Constraints
The corpus is built on journalistic standards regarding attribution and public record.
Sources: ETHICS.md:10-20
- Public Record: Everything committed to the GitHub repository is considered on-the-record and publishable.
- Source Protection: Standard attribution levels (On the record, On background, Deep background) govern what is allowed in the vault.
- Off the Record: Ephemeral information that must never be logged, stored, or referenced within the corpus.
- Privacy Principle: Agents are prohibited from creating or hydrating notes on private individuals (sources or contacts) that could identify them.
Sources: VAULT-CONVENTIONS.from-pushable-2026-05-08.md:315-325, ETHICS.md:25-45
Summary
The Idaho Journalism Corpus serves as the central memory for institutional reporting on Idaho power hierarchies. By utilizing a metadata-driven workflow and strict entity taxonomy, it enables Logan Finney to manage thousands of complex legislative and political data points with AI assistance while maintaining human accountability and source provenance.
Sources: Claude vault project memory 3-28-2026.md, IDAHO-VAULT — END-TO-END JOURNALISM WORKFLOW (AI-ASSISTED, LOCAL-FIRST)v0.1.md:235-245