Extraction Judge.Prompt — Development skill for Claude Code
Extraction Judge — Fact Durability Rubric.
How to install Extraction Judge.Prompt
Installs to ~/.claude/skills/joshuaswarren-remnic-extraction_judge.prompt/SKILL.md
mkdir -p ~/.claude/skills/joshuaswarren-remnic-extraction_judge.prompt && curl -fsSL https://raw.githubusercontent.com/joshuaswarren/remnic/HEAD/prompts/extraction_judge.prompt.md -o ~/.claude/skills/joshuaswarren-remnic-extraction_judge.prompt/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Extraction Judge.Prompt does
Extraction Judge — Fact Durability Rubric
You are a memory curator evaluating whether extracted facts are **durable** — worth storing for long-term recall across sessions.
Durability Criteria
A fact is **durable** if it will still be useful **30+ days from now** and is relevant **across multiple sessions**, not just the current task.
DURABLE examples (accept)
- Personal preferences, identities, or relationships
- Decisions with rationale that affect future work
- Corrections to prev
Alternatives in Development
- 01 Extract Strategy — Strategy Extraction Prompt 584 ★
- A B Copy — Use when you want a marketing-copy A/B — spawns two writer sibling sessions via CCC that draft the SAME messag 130 ★
- Quant Router Judge — This just a example for prompt 116 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Default Extraction
You are extracting structured facts from a document in a personal knowledge base. Extract every discrete, atom
Lens
Review a prompt, page, component, or draft through expert lenses (findings), fix it in place with --fix, or gr
Openprism Bakeoff
Blind-judge two models on one prompt to decide which earns a panel seat
Sufficient Context Judge
Sufficient-context judge prompt (Stage 2c, primair)
Analyze User Chats
Claude Code Collaboration Insights Extraction Prompt
Analyze User Chats2
Claude Code Collaboration Insights Extraction Prompt
Related Agents
Routed Judge
Judgment-class dispatch target — architecture decisions, synthesis across sources, adversarial verification, s
Mathodology Award Judge
Use for one independent blind judge seat scoring an award-level submission against contest rubric criteria.
Voice Call Judge Escalated
High-effort voice-call rubric judge for conflicting or incident evidence