curating-memories

curating-memories is a skill for Claude Code from rjmurillo/ai-agents. It costs 91 tokens per session (1,740 once invoked), scanned A, original, MIT.

A guide for keeping Forgetful memory records accurate and connected, including updating records, marking outdated information as obsolete, and linking related knowledge. It also explains in-file Serena markers for superseded content.

In plain words
What is it for?
Use it to correct memories, mark obsolete information, link related records, deduplicate entries, and clean up outdated knowledge.
Why use it?
It prevents stale or duplicated records from polluting searches and making stored knowledge harder to use.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT variable. Also seen: reads .claude/ paths.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the project-toolkit plugin — 113 skills, 26 commands, 33 agents, 4 hooks shipped together

Good fit Use it to correct memories, mark obsolete information, link related records, deduplicate entries, and clean up outdated knowledge.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add rjmurillo/ai-agents
Claude Code
/plugin install project-toolkit

Made for: Claude Code.

Or install project-toolkit, the plugin that ships this one along with the rest of its 113 skills, 26 commands, 33 agents, 4 hooks.

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 curating-memories

README.md
[![agentmods](https://agentmods.dev/badge/skills/rjmurillo/ai-agents/curating-memories/github.svg)](https://agentmods.dev/skills/rjmurillo/ai-agents/curating-memories)
Your own site
<a href="https://agentmods.dev/skills/rjmurillo/ai-agents/curating-memories"><img src="https://agentmods.dev/badge/skills/rjmurillo/ai-agents/curating-memories/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 curating-memories

Your own site · 80×15
<a href="https://agentmods.dev/skills/rjmurillo/ai-agents/curating-memories"><img src="https://agentmods.dev/badge/skills/rjmurillo/ai-agents/curating-memories.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 91 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,740 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 $0.00091 $0.01740
Opus 5 $0.00046 $0.00870
Sonnet 5 $0.00018 $0.00348
Haiku 4.5 $0.00009 $0.00174

Measured 4d ago against content hash 8052678844b6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

curating-memories 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 4d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/supersession_sweep.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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/curating-memories/SKILL.md · 145 lines

How it starts

The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Curating Memories

Active curation keeps the knowledge base accurate. A memory that is fully resolved or historical but still reads as live guidance forces a fresh agent to read past archaeology to reach current truth, and outdated content pollutes every search that touches it.

This skill owns in-file supersession markers on .serena/memories/**. It never merges or deletes Serena files and never edits Serena index files. Use memory-consolidate for cross-file Serena merges, deletions, and index cleanup, and memory-maintenance for health, token, and size checks.

Triggers

Trigger Phrase Operation
how do I mark a memory obsolete Strike the obsolete content, add a dated banner
how do I supersede a memory Apply the healthy-supersession shape below
how do I clean up stale memories Run the sweep, then verify each proposal
run the supersession sweep Scripts section below

Scope

Serena memories are markdown files, not records in a store. There is no soft-delete, no obsolete flag, and no link table to write. Curation is an edit to the file's own text, and history survives because the file keeps it visible rather than because a backend hides it.

In scope Out of scope
Strike-through and dated banners inside one memory file Merging two Serena memory files (memory-consolidate)
Collapsing a resolved investigation to a changelog footer Deleting a memory file (memory-consolidate)
Adding a dated-snapshot banner to a point-in-time doc Editing memory-index.md (memory-consolidate)
Running and verifying the supersession sweep Store health, token, and size checks (memory-maintenance)

When to Supersede

Supersede content when:

  • Newer information contradicts it.
  • A decision has been reversed or replaced.
  • The code, script, workflow, or artifact it references no longer exists.
  • The file records a resolved investigation but still reads as an open one.
  • A dated point-in-time snapshot is framed as current state.

Read the full file on GitHub · 145 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. 4d ago Changed · -106 lines · +2 tokens per session 8052678844b6
  2. 6d ago Changed · -5 lines 9dab6bdb2318
  3. 8d ago First seen · 256 lines · 89 tokens per session scan A 9ba83920573d

Subscribe to this mod's changes

curating-memories is a skill published in the GitHub repository rjmurillo/ai-agents (45 stars, last pushed today), licensed MIT. It adds 91 tokens to every session and 1,740 once invoked, about $0.0005 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 2026-09-03.