memory

Save valuable insights from the current session to storage.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/luanmorenommaciel/agentspec/memory
Clone the repo
git clone --depth 1 https://github.com/luanmorenommaciel/agentspec

Made for: Claude Code.

Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 685 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00011 $0.00685
Opus 5 $0.00005 $0.00342
Sonnet 5 $0.00002 $0.00137
Haiku 4.5 $0.00001 $0.00068

Measured yesterday against content hash 6e892f139581, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

memory 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 yesterday.

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/commands/core/memory.md · 136 lines

How it starts

The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Memory Command

Save session insights to .claude/storage/ for future reference (directory is auto-created on first use).

Usage

/memory                           # Save current session insights
/memory "specific note to save"   # Save with specific context

What It Does

  1. Analyzes current conversation for valuable insights
  2. Compresses to high-signal format (decisions, patterns, gotchas)
  3. Saves to .claude/storage/memory-{YYYY-MM-DD}.md

When to Use

Use /memory when you've discovered something worth remembering:

  • ✅ Non-obvious decisions with rationale
  • ✅ Patterns that worked well
  • ✅ Gotchas discovered
  • ✅ Architecture decisions
  • ✅ Terminology clarifications

Don't save:

  • ❌ Step-by-step implementation details (obvious from code)
  • ❌ Temporary debugging info
  • ❌ Every session (only valuable ones)

Output Format

Creates: .claude/storage/memory-{YYYY-MM-DD}.md

# Memory: {date}

> {One-line summary of session}

## Decisions Made

| Decision | Rationale | Impact |
| -------- | --------- | ------ |
| {what} | {why} | {files affected} |

## Patterns Discovered

- {pattern}: {where applied}

## Gotchas

- {gotcha}: {how to avoid}

## Open Items

- [ ] {item for next session}

---
*Saved: {timestamp}*

Process

When invoked:

1. Scan conversation for:
   - Decisions (look for "decided", "chose", "will use")
   - Patterns (look for reusable solutions)
   - Gotchas (look for "gotcha", "watch out", "careful")
   - Open items (look for "TODO", "later", "next time")

2. Compress ruthlessly:
   - Max 5 decisions
   - Max 3 patterns
   - Max 3 gotchas
   - Max 3 open items

3. Write to storage:
   - Create .claude/storage/ if not exists
   - Append to existing file if same date
   - Use consistent format

Example

User: /memory "Completed authentication refactoring"

→ Scanning conversation...
→ Found: 2 decisions, 1 pattern, 1 gotcha

Saved to: .claude/storage/memory-2026-01-23.md

## Preview:
> Completed authentication refactoring with JWT + refresh tokens

| Decision | Rationale |
| -------- | --------- |
| Use JWT + refresh | Stateless, scales better |
| 15min access token | Balance security/UX |

Pattern: Token rotation on refresh

Gotcha: Must invalidate refresh tokens on password change

Read the full file on GitHub · 136 lines

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. yesterday First seen · 136 lines · 11 tokens per session scan A 6e892f139581

Subscribe to this mod's changes

memory is a command published in the GitHub repository luanmorenommaciel/agentspec (244 stars, last pushed 2d ago), licensed MIT. It adds 11 tokens to every session and 685 once invoked, about $0.0001 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 2026-09-01.