cc-memory

A command for organising Claude Code's persistent memory across sessions. It splits project rules and learned information into separate files, with an optional memory server for larger codebases.

In plain words
What is it for?
Use it to initialise, inspect, rotate, or audit memory files, or to configure an optional long-term memory server. It can also preview changes without writing them.
Why use it?
It addresses the limits of keeping all context in one memory file, including limited space and the difficulty of finding older information.

Command for Claude Code

Part of the claude-orchestration plugin — 49 commands, 9 agents shipped together

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/thelobbi/claude/cc-memory
Clone the repo
git clone --depth 1 https://github.com/TheLobbi/claude

Made for: Claude Code.

Or install claude-orchestration, the plugin that ships this one along with the rest of its 49 commands, 9 agents.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,703 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 $0.00000 $0.02703
Opus 5 $0.00000 $0.01352
Sonnet 5 $0.00000 $0.00541
Haiku 4.5 $0.00000 $0.00270

Measured 2d ago against content hash ed8b51497257, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cc-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 2d 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/commands/cc-memory.md · 353 lines

How it starts

The opening of the file, as written. The whole thing — 353 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/cc-memory — Hybrid Memory Architecture for Claude Code

Configure persistent memory that survives across sessions using a layered approach: split rule files for always-loaded context, auto-memory for organic learning, and optional MCP-backed long-term memory for large codebases.

Usage

/cc-memory                      # Interactive setup
/cc-memory --init                # Initialize split memory architecture
/cc-memory --status              # Show memory usage, file sizes, token estimates
/cc-memory --rotate              # Archive old session summaries
/cc-memory --audit               # Check for bloat, duplicates, stale entries
/cc-memory --mcp                 # Set up MCP-backed persistent memory server
/cc-memory --dry-run             # Show what would be created without writing

The Memory Problem

Claude Code provides MEMORY.md (first 200 lines loaded per session), but this has limits:

  • 200-line cap: Only the first 200 lines are loaded — anything beyond is silently ignored
  • Single file: Everything competes for space — project identity, preferences, decisions, patterns
  • No semantic search: Claude can't query past memories, only read what's loaded
  • No session continuity: Each session starts fresh with no awareness of what happened yesterday

The Solution: 3-Tier Memory Architecture

Tier 1: Split Rule Files (Always Loaded)

Files in .claude/rules/ are auto-loaded every session with no line limit per file. Split memory by purpose instead of cramming into one MEMORY.md:

.claude/rules/
├── memory-profile.md        # Project identity, team, domains
├── memory-preferences.md    # User workflow preferences
├── memory-decisions.md      # Architecture Decision Records
├── memory-patterns.md       # Recurring code patterns & solutions
├── memory-sessions.md       # Recent session summaries (rotated)
└── lessons-learned.md       # Auto-captured errors and fixes
memory-profile.md
# Project Profile

- **Name**: {project_name}
- **Type**: {web app / API / CLI / library / monorepo}
- **Team**: {team_size, roles}
- **Domains**: {business domains}
- **Repo**: {repo_url}
- **Primary language**: {language}
- **Framework**: {framework}
- **Deploy target**: {cloud / on-prem / hybrid}

Read the full file on GitHub · 353 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. 2d ago First seen · 353 lines · 0 tokens per session scan A ed8b51497257

Subscribe to this mod's changes

cc-memory is a command published in the GitHub repository TheLobbi/claude (21 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,703 tokens. 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 2026-08-31.