Research To Json

Name: Research To Json
Author: ansanabria

ansanabria/tinto-politico-app

2 installs
1 repo stars
Updated March 2, 2026
ansanabria/tinto-politico-app

research-to-json is a skill that transforms candidate research markdown into deterministic per-candidate JSON files mapped to a Payload CMS Candidates collection.

About

research-to-json is a skill that transforms candidate research markdown files into deterministic per-candidate JSON files for seeding a Payload CMS. A developer uses it to convert, structure or normalize candidate research into one JSON file per candidate that maps directly to the Candidates collection fields. It follows explicit mapping rules, preserves Spanish text faithfully, enforces enum quality checks, and keeps output deterministic without inventing content.

Transforms candidate research markdown into deterministic per-candidate JSON
Maps headings to an explicit Candidates CMS collection contract
Enforces enum quality checks for tiers, statuses and social platforms

Research To Json by the numbers

2 all-time installs (skills.sh)
Ranked #1,839 of 2,715 Automation & Workflows skills by installs in the Skillselion catalog
Data as of Jul 28, 2026 (Skillselion catalog sync)

At a glance

research-to-json capabilities & compatibility

Capabilities: data transformation · markdown parsing · cms seeding
Use cases: data analysis · database

From the docs

What research-to-json says it does

Transform candidate research markdown files from research/ into deterministic per-candidate JSON files in data/.

SKILL.md

Keep output deterministic so future runs over the same markdown produce the same JSON shape.

SKILL.md

npx skills add https://github.com/ansanabria/tinto-politico-app --skill research-to-json

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

[![Listed on Skillselion](https://skillselion.com/badge/skills/ansanabria/tinto-politico-app/research-to-json.svg)](https://skillselion.com/skills/ansanabria/tinto-politico-app/research-to-json)

Installs	2
repo stars	★ 1
Last updated	March 2, 2026
Repository	ansanabria/tinto-politico-app ↗

What it does

Convert candidate research markdown into deterministic per-candidate JSON that maps to a Payload CMS Candidates collection.

Who is it for?

Normalizing candidate research markdown into structured, deterministic seed JSON for a CMS.

Skip if: Generating a single multi-candidate JSON document or inventing content not in the research.

When should I use this skill?

A user asks to convert, structure, normalize or prepare candidate research for seeding or Payload ingestion.

What you get

One deterministic JSON file per candidate mapped to the Candidates collection, with enum-validated fields and faithful Spanish text.

per-candidate JSON files
Candidates collection seed data

By the numbers

8 mapping rules
6 pre-write quality checks
one JSON file per candidate

Files

SKILL.mdMarkdownGitHub ↗

Research Markdown -> Candidate JSON

Use this skill to convert research/*.md files into structured candidate JSON files for seeding.

Goal

Produce one JSON file per candidate in data/:

research/ivan-cepeda.md -> data/ivan-cepeda.json
research/abelardo-de-la-espriella.md -> data/abelardo-de-la-espriella.json

Do not generate one big multi-candidate JSON document.

Output contract

Write a JSON object that maps directly to the Candidates collection fields:

{
  "name": "string",
  "slug": "string",
  "party": "string",
  "currentOffice": "string",
  "lastUpdated": "YYYY-MM-DD",
  "socialLinks": [
    { "platform": "x|instagram|facebook|youtube", "url": "https://..." }
  ],
  "biography": "string",
  "publicTrajectoryItems": [
    {
      "role": "string",
      "organization": "string",
      "startYear": "string",
      "endYear": "string",
      "location": "string",
      "description": "string"
    }
  ],
  "privateTrajectoryItems": [
    {
      "role": "string",
      "organization": "string",
      "startYear": "string",
      "endYear": "string",
      "location": "string",
      "description": "string"
    }
  ],
  "proposals": "string",
  "proposalItems": [
    {
      "title": "string",
      "description": "string",
      "topic": "string",
      "sourceTitle": "string",
      "sourceUrl": "https://...",
      "sourceTier": "oficial|prensa|ong|redes"
    }
  ],
  "controversies": "string",
  "controversyItems": [
    {
      "title": "string",
      "description": "string",
      "status": "suspicion|under_investigation|indicted|cleared|convicted",
      "year": "string",
      "sourceTitle": "string",
      "sourceUrl": "https://...",
      "sourceTier": "oficial|prensa|ong|redes"
    }
  ],
  "alliances": "string",
  "allianceParties": [{ "name": "string" }],
  "endorsers": [{ "name": "string" }],
  "record": "string",
  "funding": "string",
  "sources": [
    {
      "section": "biography|proposals|controversies|alliances|record|funding",
      "title": "string",
      "publishedAt": "YYYY-MM-DD",
      "url": "https://...",
      "tier": "oficial|prensa|ong|redes"
    }
  ],
  "summaryTrajectory": "string",
  "summaryProposals": "string",
  "summaryControversies": "string",
  "summaryAlliances": "string",
  "summaryRecord": "string",
  "summaryFunding": "string"
}

Mapping rules

1. Read the research markdown headings and map by explicit CMS field names in backticks. 2. Parse code-block list items (1., 2., etc.) into arrays preserving order. 3. Keep wording faithful to the research file; do not invent or embellish. 4. Normalize keys to exact field names from the contract above. 5. For optional arrays with no entries, use []. 6. Keep dates in YYYY-MM-DD where present. 7. Preserve Spanish text exactly (except minimal cleanup for broken whitespace). 8. Exclude operational notes (e.g., "subir manualmente") from JSON values.

Quality checks before writing

slug must match output filename.
socialLinks[].platform must be one of: x, instagram, facebook, youtube.
sourceTier and sources[].tier must be one of: oficial, prensa, ong, redes.
controversyItems[].status must be one of:

suspicion, under_investigation, indicted, cleared, convicted.

sources[].section must be one of:

biography, proposals, controversies, alliances, record, funding.

Behavior expectations

If a field is truly missing in research markdown, leave it as an empty string or empty array (whichever matches type) and mention it in your final note.
Never insert opinions, rankings, or sentiment language.
Keep output deterministic so future runs over the same markdown produce the same JSON shape.