medharness: Skill for Claude Code

.claude/skills/memory-curate/SKILL.md

memory-curate is a skill for Claude Code from charliehzm/medharness. It costs 207 tokens per session (1,974 once invoked), scanned A, original, Apache-2.0.

A maintenance tool for a project's stored knowledge. It checks factual notes, conclusions drawn from them, topic files, and workspace memory for outdated or duplicated information.

In plain words
What is it for?
Use it weekly or after major changes to update the memory index, refresh topic notes, remove duplicates, and revise conclusions affected by changed facts.
Why use it?
Without regular cleanup, an AI coding assistant may rely on old project decisions and give advice that no longer fits the codebase.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions subagents.

This is charliehzm/medharness's own configuration. It tells Claude Code how to work on medharness itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything medharness configures →

Reuse

Borrowing it

Nothing to install: this file belongs to charliehzm/medharness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/charliehzm/medharness/main/.claude/skills/memory-curate/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/charliehzm/medharness

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for memory-curate

README.md
[![agentmods](https://agentmods.dev/badge/skills/charliehzm/medharness/memory-curate/github.svg)](https://agentmods.dev/skills/charliehzm/medharness/memory-curate)
Your own site
<a href="https://agentmods.dev/skills/charliehzm/medharness/memory-curate"><img src="https://agentmods.dev/badge/skills/charliehzm/medharness/memory-curate/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for memory-curate

Your own site · 80×15
<a href="https://agentmods.dev/skills/charliehzm/medharness/memory-curate"><img src="https://agentmods.dev/badge/skills/charliehzm/medharness/memory-curate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 207 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,974 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00207 $0.01974
Opus 5 $0.00103 $0.00987
Sonnet 5 $0.00041 $0.00395
Haiku 4.5 $0.00021 $0.00197

Measured 11d ago against content hash 36758b49b812, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

memory-curate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/memory-curate/SKILL.md · 173 lines

How it starts

The opening of the file, as written. The whole thing — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Memory Curate

The least flashy and most consequential skill in the system. Without it, the project's memory layer slowly accumulates stale guidance that misleads every future AI Coding session.

Core mental model

Memory has two layers:

  • Facts — observations grounded in code / docs / decisions at a specific time
  • Inferences — derived claims about how things work / what to do

Inferences depend on facts. If a fact changes and inferences don't update, the AI starts giving advice based on a world that no longer exists.

Fact   (2026-01): "we use PostgreSQL 13, pgvector ext disabled"
Infer  (2026-01): "for vector search, use a separate Milvus instance"
   ↓
Fact   (2026-04): upgraded to PG 16, pgvector enabled        ← changed
Infer  (2026-01): "use a separate Milvus instance"            ← now stale
   ↓
Without curation: AI keeps suggesting Milvus, even though pgvector is now available
With curation:    inference is invalidated → re-derived → suggests pgvector

Your job is to detect and repair these mismatches.

What this skill produces

  1. Refreshed .memory/MEMORY.md (the index)
  2. Updated individual memory files (with bumped last_verified where applicable)
  3. .memory/curation/weekly_<YYYY-MM-DD>.md — the curation report

When NOT to use this skill

Skip for:

  • Writing new factual content (that's the producing skill's job — prd-author, openspec-apply-change, etc.)
  • Audit/compliance freezing (use audit-snapshot)
  • Project setup (no memory yet to curate)

Active context bundle

Always load first

  1. This SKILL.md
  2. .memory/MEMORY.md — the index itself
  3. reference/staleness-detection.md — rules for marking entries stale
  4. reference/merge-protocol.md — how to merge duplicates without losing nuance

Load on demand

  • reference/inference-rederivation.md — when a fact change invalidates an inference
  • reference/git-evidence-protocol.md — when using git log as staleness signal
  • All individual .memory/*.md files referenced from MEMORY.md (lazy load per topic)

Read the full file on GitHub · 173 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 173 lines · 207 tokens per session scan A 36758b49b812

Subscribe to this mod's changes

memory-curate is a skill published in the GitHub repository charliehzm/medharness (86 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 207 tokens to every session and 1,974 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories