Memory discovery
Presents memories from the default Claude Code projects location as cards, ordered oldest first, with project and memory details available for inspection.
SCMD
SCMD is a local browser-based review tool for persistent Claude Code memories. It helps Claude Code users inspect, filter, edit, delete, and restore memory entries while keeping changes staged until confirmation.
Local control over Claude Code memory
SCMD is a local review tool for persistent Claude Code memories. It loads memories from the Claude Code projects directory into a browser-based review deck, allowing users to inspect project and memory details, search and filter entries, and decide what should be kept, deleted, skipped, or edited.
The workflow separates review from modification: decisions remain staged until the user confirms Apply. Accepted edits update the memory and its matching `MEMORY.md` entry, while accepted deletions move memory files to local trash with metadata for later restoration. SCMD binds its server to `127.0.0.1`, uses a per-launch request token, and does not send data to an SCMD service.
Core capabilities
Search and filtering
Filters by project or memory type and searches memory text, including read-only `CLAUDE.md` instructions, so relevant entries can be reviewed together.
Staged decisions
Keeps keep, delete, skip, and undo decisions staged in the browser instead of modifying files during the review pass.
Controlled apply
Applies a confirmed batch while moving deleted memories to restorable trash and refusing to overwrite a memory that changed after it was loaded.
Origin and rewrite context
Can show the session message associated with a memory when its origin is available, and can request a rewrite through the local Claude Code CLI or accept a hand edit before applying it.
Local review server
Runs as a local server with live updates for decisions and file-state changes, uses no runtime dependencies, and protects API requests with a per-launch token.
Practical uses
Routine memory cleanup
Review accumulated memories under a project before they continue influencing later Claude Code sessions. Search the memory text, inspect project context, and stage keep, skip, delete, or edit decisions for one confirmed batch.
Trace questionable context
Investigate why a particular memory exists by viewing its originating session message when that source can be found, then decide whether the saved context is still appropriate.
Correct or rewrite memories
Refine stale or poorly worded memory entries without editing files immediately. Request a rewrite through the local Claude Code CLI or make a manual edit, inspect the staged result, and apply it only after confirmation.
Restore reviewed deletions
Recover from an accepted deletion when a memory is needed again. Use the explicit Restore action to put the memory and its matching `MEMORY.md` line back, or use Purge for a separate permanent deletion of a selected trash run.
Pros and Cons
Pros
- Separates review decisions from file changes until an explicit Apply action.
- Provides restorable trash for accepted memory deletions and protects against overwriting memories changed after loading.
- Offers search, project and memory-type filters, origin context, and live file-state updates in one local review workflow.
- Uses a localhost server, per-launch API token, and no SCMD telemetry or SCMD service for the review data.
Cons
- The tool is focused on Claude Code memory locations and workflows rather than general-purpose file or knowledge-base management.
- Node.js 18 or newer is required, and Windows use currently requires Git Bash for the documented Claude Code launch path.
- The available documentation does not establish team collaboration, hosted access, or multi-user support.
FAQ
What is required to run SCMD?
Install Node.js 18 or newer, including npm and npx. The primary launch path is the Claude Code plugin; a terminal launch is also available through npx. Windows use currently requires Git Bash for the plugin command and shell snippets.
How do I start SCMD from Claude Code?
In Claude Code, add the marketplace with `/plugin marketplace add nelsonjohnsolo/scmd`, install it with `/plugin install scmd@scmd`, reload plugins, and run `/scmd:run`. The plugin starts SCMD in the background and reports the local review URL in the session.
What data does SCMD review?
By default, SCMD reads memory files and indexes under `~/.claude/projects`, along with relevant project session transcripts and global or project `CLAUDE.md` files. A custom Claude configuration or auto-memory location can be supplied with `--root <dir>`.
Can changes be restored after applying them?
Review decisions remain staged in the browser until Apply is confirmed. Accepted deletions move memories to `~/.scmd/trash/` with restore metadata, while accepted edits update the memory and its matching index entry. Restore and Purge are separate explicit actions.
Quick Facts
- Category
- Developer Tool
- Primary workflow
- Discover → Review → Confirm → Restore
- Designed for
- Claude Code memory review
- Runtime requirement
- Node.js 18 or newer
- Deployment model
- Local server on 127.0.0.1
- License
- MIT
Alternatives à SCMD
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
GitBot
GitBot is a local browser-based workspace for creating reusable coding bots powered by Claude Code, Codex, or OpenCode. It helps developers run those agents across repository folders with persistent threads and configurable permissions.
hob
hob is an independent workspace for coding agents that keeps agent sessions, terminals, history, and follow-up work organized around the tools and providers you already use. It is aimed at developers who want local control over routing, history, and workspace structure rather than a bundled model stack.
Ably Chat
Ably Chat is a chat API platform for building custom realtime chat applications. It supports room-based messaging, typing indicators, presence, reactions, and message updates, with usage-based pricing options for different deployment stages.
Manta AI
Manta AI is an autonomous web app testing tool for teams that want to map application behavior, catch regressions, and generate tests without writing scripts or maintaining selectors. It works from a URL and supports plain-English test flows, run results with screenshots, and scheduled or deployment-triggered checks.