Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Judge Rem Memory Updates

ASecurity

Judge REM memory relations, rewrites, verification, and clause carry by meaning. Use for REM update model calls that decide which older facts are superseded, which facts remain current, and whether a proposed rewrite preserves all surviving facts.

34 stars
0 votes
0 copies
0 views
Added 9/22/2026
ai-agents

Security Analysis

A100/100

Scanned 9/22/2026

Install to Claude Code

$npx -y skills add sno-ai/sno-station --skill judge-rem-memory-updates --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Judge Rem Memory Updates?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Judge Rem Memory Updates
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/sno-ai-judge-rem-memory-updates/badge)](https://www.skillsdirectory.com/skills/sno-ai-judge-rem-memory-updates)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: judge-rem-memory-updates
description: Judge REM memory relations, rewrites, verification, and clause carry by meaning. Use for REM update model calls that decide which older facts are superseded, which facts remain current, and whether a proposed rewrite preserves all surviving facts.
---

# Judge REM Memory Updates

Judge meaning. Do not decide from shared words, punctuation, or text position.

## Relation Judgment

- Mark `supersedes` true only when the newer record replaces meaning in the older record.
- Mark `retires_anything` true only when at least one meaning in the older record stops being current.
- Mark `supersedes_everything` true only when every current meaning in the older record is replaced.

## Source Rewrite

- Put only the facts that remain current in `proposed_current`.
- Quote each superseded value from the source row in `retired_values`.
- Drop the wording that exists only to describe the change itself — what the value was
  before, when it changed, and who changed it. That wording retires with the value it
  describes. "currently $1,100,000, updated from $900,000 on 2025-06-06" rewrites to
  "currently $1,100,000", and `retired_values` carries "$900,000".
- This holds in every language and for every kind of value, not only amounts. Any phrasing of
  the shape "changed from A to B", "shifted from A to B", "switched from A to B", "moved from A
  to B", "A is now B", "no longer A but B" leaves ONLY B in `proposed_current` and puts A in
  `retired_values`. The old value must not appear anywhere in the rewrite, not even inside the
  phrase that describes the change — copying "shifted from A to B" into `proposed_current`
  keeps the retired value alive and is wrong. Rewrite it as B alone.
  English: "shifted from morning tea to coffee" rewrites to "coffee", retiring "morning tea".
  Spanish: "cambió de té matutino a café" rewrites to "café", retiring "té matutino".
- A line that says an item is no longer needed, no longer tracked, done with, or that asks
  for it to be removed from the list, IS a retirement instruction. Drop that whole line from
  `proposed_current`, and drop the item it names as well. Quote the named item in
  `retired_values`. Leaving either one in place is wrong: the instruction line is wording
  about a change, and the item it names is the value that change retires.
- Return no retired values when nothing was superseded.
- Return an empty proposed value when nothing remains current.

## Rewrite Verification

- Mark `faithful` true only when source evidence supports every claim in the rewrite.
- When a newer record is provided, retire only meanings that the newer record replaces.
- Reject any unsupported fact, entity, quantity, or claim.
- Mark `retired_absent` true only when no retired meaning remains current in the rewrite.
- Mark `all_facts_accounted` true only when every source fact remains current or appears in the retired values.
- Wording that exists only to describe the change — the prior value, the date it changed,
  who changed it — is part of the retirement, not a surviving fact. Its absence from the
  rewrite never makes `all_facts_accounted` false.
- Accept paraphrases that keep the same meaning.

## Clause Carry

- Mark a clause already current only when the survivor asserts the same meaning as current.
- A negation, quotation, or past description is not a current assertion.
- Different words with the same meaning count as current.

Attribution

sno-aisno-ai
View sourceMore from sno-ai →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1066601 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →