neuralscape

Ingest files, folders, or a block of pasted context into Neuralscape memory. Use when the user says "ingest this file/folder", "add these docs to memory", "index this PDF/Word doc", "load this documentation into Neuralscape", or pastes a long passage they want remembered. Handles Markdown, HTML, PDF, and MS Office (docx/xlsx/pptx); files are stored as artifacts and referenced back. Works in both Claude Code and Claude Cowork.

Check that Neuralscape is reachable and show the current plugin configuration. Use when the user asks "is neuralscape working", "is my memory service up", "what URL is neuralscape using", or any health/connectivity question. Works in both Claude Code and Claude Cowork.

Load the user's relevant Neuralscape memories before planning or acting — their preferences, conventions, tech stack, and past decisions. Use at the start of a task, when the user says "neuralscape on" / "load my context" / "what do you know about me or this project", or whenever you need prior context. Works in both Claude Code and Claude Cowork (MCP-driven).

Extract and save the durable facts from the current conversation to Neuralscape at the end of a substantive working session. Use when the user says "save this session", "save this conversation", "compile this session", or when wrapping up meaningful work. Works in both Claude Code and Claude Cowork (MCP-driven) — and is the Cowork stand-in for the Claude Code Stop hook.

Search the user's Neuralscape memory store for relevant facts. Use when the user asks "what do I know about X", "have we discussed Y", "what did I decide about Z", or any retrieval-style question. Returns hybrid vector + graph results. Works in both Claude Code and Claude Cowork (MCP-driven).

Save one durable fact to Neuralscape memory now — a preference, decision, convention, gotcha, or architectural choice. Use when the user says "remember this", "save that", "note that I prefer X", or when you learn something worth keeping for future sessions. Works in both Claude Code and Claude Cowork (MCP-driven).

Read a JSONL buffer of captured tool observations and extract significant memories using the Neuralscape memory model v2. Use this when a UserPromptSubmit, SessionStart, or `/neuralscape:capture` instruction asks you to compile observations from a buffer file path.

Manually flush the current conversation to Neuralscape's compiler. Use after important sessions when the user wants context captured immediately, or when they say "sync my conversation", "save this to memory", or "compile this session now".

Start a defined team process by describing what you want to do. Run this when the user types "/process ...", says "start a process", "run the <X> process", or describes a task that matches a standard team workflow (e.g. "build the POV deck for Acme", "draft the follow-up email", "research this account"). The skill matches the request to one of the org's predefined processes, pulls that process's authoritative standards, and steers the session to run it. MCP-driven — works in Claude Code and Claude Cowork.

Manually compile the current session's PostToolUse observation buffer into Neuralscape memories. Use when the user types `/neuralscape:capture` or asks to "flush memories now" / "save what we've done so far".

Show the current Neuralscape plugin configuration (URL, user_id, API-key state) and how identity is resolved on this platform. Use when the user asks "what's my neuralscape config", "where is neuralscape pointing", "am I logged in to neuralscape", or wants to verify their settings. Works in both Claude Code and Claude Cowork.

List the user's Neuralscape projects and pick (or create) the one to scope memory to for this session. Use when the user says "switch project", "what projects do I have", "set project to X", or when a recall/remember needs a project and none is active. Especially useful in Claude Cowork, which has no working directory to infer a project from. MCP-driven — works in both platforms.

Use the Neuralscape memory layer to recall and store memories about the user and their projects. Use this at session start to load context, when you learn new facts, and when the user asks about preferences, conventions, or past decisions. Works via MCP tools or REST API at localhost:8199. Memory writes are async (processed by background workers via ARQ + Redis). Multi-user: each authenticated user has a private memory pool plus a shared team pool — pick the right visibility per write.

Add new agent framework adapters to NeuralScape's unified conversation hook system. Use when integrating a new AI client (Cursor, Copilot, custom agent), adding a client adapter, or extending the plugin to support another agent framework. Covers adapter creation, client detection, hook manifest, and build config.