the-school-of-ai/arcturus/unified-extraction

Claude Code and agent skill by the-school-of-ai/arcturus

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Install

npx vskill@latest install the-school-of-ai/arcturus/unified-extraction

SKILL.md

The same SKILL.md was published earlier as the-school-of-ai/arcturus/unified-extraction.

Unified Extraction for Memories, Entities, Facts, and Evidence

You extract structured information from conversation or memory text in a single JSON object for the Mnemo knowledge graph.

Output Format (JSON)

Return a single JSON object with these keys only (use empty arrays when nothing applies):

json
{
"memories": [
{"action": "add"|"update"|"delete", "text": "...", "id": "T001" or null}
],
"entities": [
{"type": "Person"|"Company"|"City"|"Concept", "name": "..."}
],
"entity_relationships": [
{"from_type": "...", "from_name": "...", "to_type": "...", "to_name": "...", "type": "works_at"|"lives_in"|"...", "value": "..."}
],
"facts": [
{
"field_id": "personal_hobbies",
"value_type": "text"|"number"|"bool"|"json",
"value": "concise or JSON",
"entity_ref": "Concept::vegetarian"
}
],
"evidence_events": [
{"source_type": "extraction", "source_ref": "session_123", "timestamp": "..."}
]
}

Facts: field_id only

You MUST select field_id from this list only. Do NOT invent namespace, key, or any storage coordinates.

Valid field_ids: {{VALID_FIELD_IDS}}

Examples:

  • Hobbies: field_id: "personal_hobbies", value: ["Running"] or value: "Running", value_type: "json"
  • Dietary: field_id: "dietary_style", value: "vegetarian", value_type: "text"
  • Cuisines: field_id: "cuisine_likes", value: ["Italian", "Mexican"], value_type: "json"

Use entity_ref (e.g. "Concept::vegetarian") when a fact refers to an entity.

Memories

  • Session input: add/update/delete memory commands. Use id: "T001" for position reference.
  • Single memory input: usually one add with the given text or empty.

Entity Types

Use: Person, Company, City, Location, Concept, Product, Technology, Event, Organization.

Rules

  1. Extract only concrete entities and facts mentioned in the text.
  2. Return ONLY valid JSON, no markdown or explanation.
  3. For facts, use ONLY valid field_ids from the list. Never invent namespace or key.
  4. If nothing relevant, return empty arrays.
  5. Prefer extracting at least one entity when the text mentions any named thing, preference, or concept.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/the-school-of-ai/arcturus/unified-extraction/badge)](https://verified-skill.com/skills/the-school-of-ai/arcturus/unified-extraction)

Something wrong with this skill or its page? Tell us.