🧠 Prism MCP — The Mind Palace for AI Agents

Your AI agent forgets everything between sessions. Prism fixes that — then teaches it to think.

Prism v9.0 is the Autonomous Cognitive OS for AI agents. Built on a true cognitive architecture inspired by human brain mechanics, Prism gives agents affect-tagged memory, token-economic cost awareness, and multi-hop graph reasoning — your agent now follows causal trains of thought across memory, forms principles from experience, economizes what it stores, and knows when it lacks information. Your agents don't just remember; they think.

npx -y prism-mcp-server

Works with Claude Desktop · Claude Code · Cursor · Windsurf · Cline · Gemini · Antigravity — any MCP client.

Table of Contents

• Why Prism?

• Quick Start

• The Magic Moment

• Setup Guides

• Universal Import: Bring Your History

• What Makes Prism Different

• Autonomous Cognitive OS (v9.0)

• Synapse Engine (v8.0)

• Cognitive Architecture (v7.8)

• Data Privacy & Egress

• Use Cases

• What's New

• How Prism Compares

• Tool Reference

• Environment Variables

• Architecture

• Scientific Foundation

• Milestones & Roadmap

• Enterprise & Commercial Support

• Troubleshooting FAQ

Why Prism?

Every time you start a new conversation with an AI coding assistant, it starts from scratch. You re-explain your architecture, re-describe your decisions, re-list your TODOs. Hours of context — gone.

Prism gives your agent a brain that persists — and then teaches it to reason. Save what matters at the end of each session. Load it back instantly on the next one. But Prism goes far beyond storage: it consolidates raw experience into lasting principles, traverses causal chains to surface root causes, and knows when to say "I don't know."

📌 Terminology: Throughout this doc, "Prism" refers to the MCP server and cognitive memory engine. "Mind Palace" refers to the visual dashboard UI at localhost:3000 — your window into the agent's brain. They work together; the dashboard is optional.

Prism has three pillars:

1. 🧠 Cognitive Memory — Memories are ranked like a human brain: recently and frequently accessed context surfaces first, while stale context fades naturally via ACT-R activation decay. Raw experience consolidates into semantic principles through Hebbian learning. The result is retrieval quality that no flat vector search can match. (See Cognitive Architecture and Scientific Foundation.)

2. ⚡ Synapse Engine (GraphRAG) — When your agent searches for "Error X", the Synapse Engine doesn't just find logs mentioning "Error X". Multi-hop energy propagation traverses the causal graph — dampened by fan effect, bounded by lateral inhibition — and surfaces "Workaround Y" connected to "Architecture Decision Z". Nodes discovered exclusively via graph traversal are tagged [🌐 Synapse] so you can see the engine working. (See Synapse Engine.)

3. 🏭 Autonomous Execution (Dark Factory) — When you're ready, Prism can run coding tasks end-to-end with a fail-closed pipeline where an adversarial evaluator catches bugs the generator missed — before you ever see the PR. (See Dark Factory.)

Quick Start

Prerequisites

• Node.js v18+ (v20 LTS recommended; v23.x has known npx quirk)

• Any MCP-compatible client (Claude Desktop, Cursor, Windsurf, Cline, etc.)

• No API keys required for core features (see Capability Matrix)

Install

Add to your MCP client config (claude_desktop_config.json, .cursor/mcp.json, etc.):

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"]
    }
  }
}

⚠️ Windows / Restricted Shells: If your MCP client complains that npx is not found, use the absolute path to your node binary (e.g. C:\Program Files\nodejs\npx.cmd).

That's it. Restart your client. All tools are available. The Mind Palace Dashboard (the visual UI for your agent's brain) starts automatically at http://localhost:3000. You don't need to keep a tab open — the dashboard runs in the background and the MCP tools work with or without it.

🔮 Pro Tip: Once installed, open http://localhost:3000 in your browser to view the Mind Palace Dashboard — a beautiful, real-time UI of your agent's brain. Explore the Knowledge Graph, Intent Health gauges, and Session Ledger.

🔄 Updating Prism: npx -y caches the package locally. To force an update to the latest version, restart your MCP client — npx -y will fetch the newest release automatically. If you're stuck on a stale version, run npx clear-npx-cache (or npm cache clean --force) before restarting.

Add PRISM_DASHBOARD_PORT to your MCP config env block:

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"],
      "env": { "PRISM_DASHBOARD_PORT": "3001" }
    }
  }
}

Then open http://localhost:3001 instead.

Capability Matrix

🔑 The core Mind Palace works 100% offline with zero API keys. Cloud keys unlock intelligence features. See Environment Variables.

💰 API Cost Note: GOOGLE_API_KEY (Gemini) has a generous free tier that covers most individual use. BRAVE_API_KEY offers 2,000 free searches/month. FIRECRAWL_API_KEY has a free plan with 500 credits. For typical solo development, expect $0/month on the free tiers. Only high-volume teams or heavy autonomous pipeline usage will incur meaningful costs.

The Magic Moment

Session 1 (Monday evening):

You: "Analyze this auth architecture and plan the OAuth migration."
Agent: *deep analysis, decisions, TODO list*
Agent: session_save_ledger → session_save_handoff ✅

Session 2 (Tuesday morning — new conversation, new context window):

Agent: session_load_context → "Welcome back! Yesterday we decided to use PKCE
       flow with refresh tokens. 3 TODOs remain: migrate the user table,
       update the middleware, and write integration tests."
You: "Pick up where we left off."

Your agent remembers everything. No re-uploading files. No re-explaining decisions.

Setup Guides

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"]
    }
  }
}

Add to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"]
    }
  }
}

Add to your Continue config.json or Cline MCP settings:

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"],
      "env": {
        "PRISM_STORAGE": "local",
        "BRAVE_API_KEY": "your-brave-api-key"
      }
    }
  }
}

Claude Code naturally picks up MCP tools by adding them to your workspace .clauderules. Simply add:

Always start the conversation by calling `mcp__prism-mcp__session_load_context(project='my-project', level='deep')`.
When wrapping up, always call `mcp__prism-mcp__session_save_ledger` and `mcp__prism-mcp__session_save_handoff`.

Format Note: Claude automatically wraps MCP tools with double underscores (mcp__prism-mcp__...), while most other clients use single underscores (mcp_prism-mcp_...). Prism's backend natively handles both formats seamlessly.

See the Gemini Setup Guide for the proven three-layer prompt architecture to ensure reliable session auto-loading.

To sync memory across machines or teams:

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"],
      "env": {
        "PRISM_STORAGE": "supabase",
        "SUPABASE_URL": "https://your-project.supabase.co",
        "SUPABASE_KEY": "your-supabase-anon-or-service-key"
      }
    }
  }
}

Schema Migrations

Prism auto-applies its schema on first connect — no manual step required. If you need to apply or re-apply migrations manually (e.g. for a fresh project or after a version bump), run the SQL files in supabase/migrations/ in numbered order via the Supabase SQL Editor or the CLI:

# Via CLI (requires supabase CLI + project linked)
supabase db push

# Or apply a single migration via the Supabase dashboard SQL Editor
# Paste the contents of supabase/migrations/0NN_*.sql and click Run

Key migrations:

• 020_* — Core schema (ledger, handoff, FTS, TTL, CRDT)

• 033_memory_links.sql — Associative Memory Graph (MemoryLinks) — required for session_backfill_links

Anon key vs. service role key: The anon key works for personal use (Supabase RLS policies apply). Use the service role key for team deployments where multiple users share the same Supabase project — it bypasses RLS and allows Prism to manage all rows regardless of auth context. Never expose the service role key client-side.

git clone https://github.com/dcostenco/prism-mcp.git
cd prism-mcp && npm install && npm run build

Then add to your MCP config:

{
  "mcpServers": {
    "prism-mcp": {
      "command": "node",
      "args": ["/path/to/prism-mcp/dist/server.js"],
      "env": {
        "BRAVE_API_KEY": "your-key",
        "GOOGLE_API_KEY": "your-gemini-key"
      }
    }
  }
}

Prism can be deployed natively to cloud platforms like Render so your agent's memory is always online and accessible across different machines or teams.

1. Fork this repository.

2. In the Render Dashboard, create a new Web Service pointing to your repository.

3. In the setup wizard, select Docker as the Runtime.

4. Set the Dockerfile path to Dockerfile.smithery.

5. Connect your local MCP client to your new cloud endpoint using the sse transport:

{
  "mcpServers": {
    "prism-mcp-cloud": {
      "command": "npx",
      "args": ["-y", "supergateway", "--url", "https://your-prism-app.onrender.com/sse"]
    }
  }
}

Note: The Dockerfile.smithery uses an optimized multi-stage build that compiles Typescript safely in a development environment before booting the server in a stripped-down production image. No NPM publishing required!

Common Installation Pitfalls

❌ Don't use npm install -g: Hardcoding the binary path (e.g. /opt/homebrew/Cellar/node/23.x/bin/prism-mcp-server) is tied to a specific Node.js version — when Node updates, the path silently breaks.

✅ Always use npx instead:

{
  "mcpServers": {
    "prism-mcp": {
      "command": "npx",
      "args": ["-y", "prism-mcp-server"]
    }
  }
}

npx resolves the correct binary automatically, always fetches the latest version, and works identically on macOS, Linux, and Windows. Already installed globally? Run npm uninstall -g prism-mcp-server first.

❓ Seeing warnings about missing API keys on startup? That's expected and not an error. BRAVE_API_KEY / GOOGLE_API_KEY warnings are informational only — core session memory works with zero keys. See Environment Variables for what each key unlocks.

💡 Do agents auto-load Prism? Agents using Cursor, Windsurf, or other MCP clients will see the session_load_context tool automatically, but may not call it unprompted. Add this to your project's .cursorrules (or equivalent system prompt) to guarantee auto-load:

At the start of every conversation, call session_load_context with project "my-project" before doing any work.

Claude Code users can use the .clauderules auto-load hook shown in the Setup Guides. Prism also has a server-side fallback (v5.2.1+) that auto-pushes context after 10 seconds if no load is detected.

Universal Import: Bring Your History

Switching to Prism? Don't leave months of AI session history behind. Prism can ingest historical sessions from Claude Code, Gemini, and OpenAI and give your Mind Palace an instant head start — no manual re-entry required.

Import via the CLI or directly from the Mind Palace Dashboard (Import tab → file picker + dry-run toggle).

Supported Formats

• Claude Code (.jsonl logs) — Automatically handles streaming chunk deduplication and requestId normalization.

• Gemini (JSON history arrays) — Supports large-file streaming for 100MB+ exports.

• OpenAI (JSON chat completion history) — Normalizes disparate tool-call structures into the unified Ledger schema.

How to Import

Option 1 — CLI:

# Ingest Claude Code history
npx -y prism-mcp-server universal-import --format claude --path ~/path/to/claude_log.jsonl --project my-project

# Dry run (verify mapping without saving)
npx -y prism-mcp-server universal-import --format gemini --path ./gemini_history.json --dry-run

Option 2 — Dashboard: Open localhost:3000, navigate to the Import tab, select the format and file, and click Import. Supports dry-run preview.

Why It's Safe to Re-Run

• Memory-Safe Streaming: Processes massive log files line-by-line using stream-json to prevent Out-of-Memory (OOM) crashes.

• Idempotent Dedup: Content-hash prevents duplicate imports on re-run (skipCount reported).

• Chronological Integrity: Uses timestamp fallbacks and requestId sorting to preserve your memory timeline.

• Smart Context Mapping: Extracts cwd, gitBranch, and tool usage patterns into searchable metadata.

What Makes Prism Different

💰 Token-Economic Memory & Surprisal Gates (v9.0)

Standard agents dump repetitive garbage into vector databases until they bankrupt your API budget and choke their own context windows. Prism introduces a Persistent Cognitive Budget with Universal Basic Income (UBI). Before saving, Prism calculates the semantic novelty (Surprisal) of the thought. Redundant boilerplate is penalized with a 2× cost multiplier; novel insights are rewarded. The agent is mathematically forced to learn compression.

🎭 Affect-Tagged Emotional Routing (v9.0)

Standard RAG retrieves documents based on similarity, not success rate. Prism tags experience events with Valence (-1.0 to +1.0). Highly emotional memories (catastrophic failures or massive successes) receive an affective salience boost during retrieval. If an agent heads down an architectural path that previously caused outages, Prism injects a preemptive ⚠️ Historical Friction warning, giving your AI a "gut feeling" about bad code.

🧠 Your Agent Learns From Mistakes

When you correct your agent, Prism tracks it. Corrections accumulate importance over time. High-importance lessons auto-surface as warnings in future sessions — and can even sync to your .cursorrules file for permanent enforcement. Your agent literally gets smarter the more you use it.

🕰️ Time Travel

Every save creates a versioned snapshot. Made a mistake? memory_checkout reverts your agent's memory to any previous state — like git revert for your agent's brain. Full version history with optimistic concurrency control.

🔮 Mind Palace Dashboard

A gorgeous glassmorphism UI at localhost:3000 that lets you see exactly what your agent is thinking:

• Current State & TODOs — the exact context injected into the LLM's prompt

• Intent Health Gauges — per-project 0–100 health score with staleness decay, TODO load, and decision signals

• Interactive Knowledge Graph — force-directed neural graph with click-to-filter, node renaming, and surgical keyword deletion

• Deep Storage Manager — preview and execute vector purge operations with dry-run safety

• Session Ledger — full audit trail of every decision your agent has made

• Time Travel Timeline — browse and revert any historical handoff version

• Visual Memory Vault — browse VLM-captioned screenshots and auto-captured HTML states

• Hivemind Radar — real-time active agent roster with role, task, and heartbeat

• Morning Briefing — AI-synthesized action plan after 4+ hours away

• Brain Health — memory integrity scan with one-click auto-repair

🧬 10× Memory Compression

Powered by a pure TypeScript port of Google's TurboQuant (inspired by Google's ICLR research), Prism compresses 768-dim embeddings from 3,072 bytes → ~400 bytes — enabling decades of session history on a standard laptop. No native modules. No vector database required.

🐝 Multi-Agent Hivemind

Multiple agents (dev, QA, PM) can work on the same project with role-isolated memory. Agents discover each other automatically, share context in real-time via Telepathy sync, and see a team roster during context loading. → Multi-agent setup example

🚦 Task Router

Prism can score coding tasks and recommend whether to keep execution on the host model or delegate to a local Claw agent (a lightweight sub-agent powered by Ollama/vLLM for fast, local-safe edits). This enables faster handling of small edits while preserving host execution for complex work. In client startup/skill flows, use defensive delegation: route only coding tasks, call session_task_route only when available, delegate to claw only when executor tooling exists and task is non-destructive, and fallback to host when router/executor is unavailable. → Task router real-life example

🖼️ Visual Memory

Save UI screenshots, architecture diagrams, and bug states to a searchable vault. Images are auto-captioned by a VLM (Claude Vision / GPT-4V / Gemini) and become semantically searchable across sessions.

🔭 Full Observability

OpenTelemetry spans for every MCP tool call, LLM hop, and background worker. Route to Jaeger, Grafana, or any OTLP collector. Configure in the dashboard — zero code changes.

🌐 Autonomous Web Scholar

Prism researches while you sleep. A background pipeline searches the web, scrapes articles, synthesizes findings via LLM, and injects results directly into your semantic memory — fully searchable on your next session. Brave Search → Firecrawl scrape → LLM synthesis → Prism ledger. Task-aware, Hivemind-integrated, and zero-config when API keys are missing (falls back to Yahoo + Readability).

Dark Factory — Adversarial Autonomous Pipelines

When you trigger a Dark Factory pipeline, Prism doesn't just run your task — it fights itself to produce high-quality output. A PLAN_CONTRACT step locks a machine-parseable rubric before any code is written. After execution, an Adversarial Evaluator (in a fully isolated context) scores the output against the rubric. It cannot pass the Generator without providing exact file and line evidence for every failing criterion. Failed evaluations inject the critique directly into the Generator's retry prompt so it's never flying blind. The result: security issues, regressions, and lazy debug logs caught autonomously — before you ever see the PR. → See it in action

🤖 Autonomous Cognitive OS (v9.0)

Memory isn't just about storing data; it's about economics and emotion. v9.0 transforms Prism from a passive memory database into a living Cognitive Operating System that forces agents to learn compression and develop intuition.

Most AI agents have an infinite memory budget. They dump massive, repetitive logs into vector databases until the context window chokes. Prism v9.0 fixes this by introducing Token-Economic Reinforcement Learning and Affect-Tagged Memory.

1. Memory-as-an-Economy (The Surprisal Gate)

Prism gives your project a strict Cognitive Budget (e.g., 2,000 tokens). Every time the agent saves a memory, it costs tokens.

But not all memories are priced equally:

• Prism intercepts the save and runs a Vector-Based Surprisal calculation against recent memories.

• High Surprisal (Novel thought): Costs 0.5× tokens. The agent is rewarded for new insights.

• Low Surprisal (Boilerplate): Costs 2.0× tokens. The agent is penalized for repeating itself.

• Universal Basic Income (UBI): The budget recovers passively over time (+100 tokens/hour) or when the agent successfully completes autonomous evaluations.

The Result: You don't need to prompt the agent to "be concise." The physics of the system force the LLM to learn data compression to avoid bankruptcy.

2. Affect-Tagged Memory (Giving AI a "Gut Feeling")

Vector math measures semantic similarity, not sentiment. If an agent searches for "Authentication Architecture," standard RAG will return two past approaches—it doesn't know that Approach A caused a 3-day production outage, while Approach B worked perfectly.

• Affective Salience: Prism automatically tags experience events with a valence score (-1.0 for failure, +1.0 for success).

• Emotional Retrieval: At retrieval time, the absolute magnitude (|valence|) significantly boosts the memory's ranking score. Extreme failures and extreme successes surface to the top.

• UX Warnings: If the retrieved memories are historically negative, Prism intercepts the prompt injection: ⚠️ Caution: This topic is strongly correlated with historical failures. Review past decisions before proceeding.

The Paradigm Shift

Synapse Engine (v8.0)

Standard RAG retrieves documents. GraphRAG traverses relationships. The Synapse Engine does both — a pure, storage-agnostic multi-hop propagation engine that turns your agent's memory into an associative reasoning network.

The Synapse Engine (v8.0) replaces the legacy SQL-coupled spreading activation with a pure functional graph propagation core inspired by ACT-R cognitive architecture. It is Prism's native, low-latency GraphRAG solution — no external graph database required.

Before vs After

How It Works

  Query: "Project Apollo status"
              │
  ┌───────────┼───────────────┐
  ▼           ▼               ▼
[Design    [Sprint          [Deployment
  Doc]      Retro]            Log]
  │ 1.0       │ 0.8            │ 0.6    ← semantic anchors
  │           │                │
  ▼           ▼                ▼
[Dev       [Auth Bug      [Perf         ← Synapse discovered
  Profile    Thread  🌐]    Regression 🌐]   (multi-hop)
  0.42]      0.38]          0.31]

Key design decisions:

💡 Synapse is non-fatal — if the graph traversal fails for any reason, search gracefully returns the original semantic matches. Zero risk of degraded search.

Cognitive Architecture (v7.8)

Prism v7.8 is our biggest leap forward yet. We have moved beyond flat vector search and implemented a true Cognitive Architecture inspired by human brain mechanics. With the new ACT-R Spreading Activation Engine, Episodic-to-Semantic memory consolidation, and Uncertainty-Aware Rejection Gates, Prism doesn't just store logs anymore — it forms principles, follows causal trains of thought, and possesses the self-awareness to know when it lacks information.

Standard RAG (Retrieval-Augmented Generation) is now a commodity. Everyone has vector search. What turns a memory storage system into a memory reasoning system is the cognitive layer between storage and retrieval. Here is what Prism v7.8 builds on top of the vector foundation:

1. The Agent Actually Learns (Episodic → Semantic Consolidation)

How it works: When Prism compacts session history, it doesn't just summarize text — it extracts principles. Raw event logs ("We deployed v2.3 and the auth service crashed because the JWT secret was rotated") consolidate into a semantic rule ("JWT secrets must be rotated before deployment, not during"). These rules live in a dedicated semantic_knowledge table with confidence scores that increase every time the pattern is observed. Your agent doesn't just remember what it did; it learns how the world works over time. This is true Hebbian learning: neurons that fire together wire together.

2. "Train of Thought" Reasoning (Spreading Activation & Causality)

How it works: When compacting memories, Prism extracts causal links (caused_by, led_to) and persists them as edges in the knowledge graph. At retrieval time, ACT-R spreading activation propagates through these edges with a damped fan effect (1 / ln(fan + e)) to prevent hub-flooding, lateral inhibition to suppress noise, and configurable hop depth. If you search for "Error X", the engine traverses the graph and brings back "Workaround Y" → "Architecture Decision Z" — a literal train of thought instead of a static search result.

  Query: "Why does the API timeout?"
                    │
      ┌─────────────┼─────────────┐
      ▼             ▼             ▼
  [Memory: API     [Memory:      [Memory:       
   timeout error]   DB pool       rate limiter
                    exhaustion]   misconfigured]
      │                │
      ▼                ▼
  [Memory:         [Memory:
   caused_by →      led_to →
   connection       connection
   leak in v2.1]    pool patch
                    in v2.2]

3. Self-Awareness & The End of Hallucinations (The Rejection Gate)

How it works: The Uncertainty-Aware Rejection Gate operates on two signals: similarity floor (is the best match even remotely relevant?) and gap distance (is there meaningful separation between the top results, or are they all equally mediocre?). When both signals indicate low confidence, Prism returns a structured rejection — telling the LLM "I searched my memory, and I confidently do not know the answer" — instead of feeding it garbage context that causes hallucinations. In the current LLM landscape, an agent that knows its own boundaries is a massive competitive advantage.

4. Block Amnesia Solved (Dynamic Fast Weight Decay)

How it works: Most memory systems decay everything at the same rate, meaning agents eventually forget their core system instructions as time passes. Prism applies ACT-R base-level activation decay (B_i = ln(Σ t_j^(-d))) with a 50% slower decay rate for semantic rollup nodes (ageModifier = 0.5 for is_rollup entries). The agent will naturally forget what it ate for breakfast (raw episodic chatter), but it will permanently remember its core personality, project rules, and hard-won architectural decisions. The result: Long-Term Context anchors that survive indefinitely.

Data Privacy & Egress

Where is my data stored?

All data lives under ~/.prism-mcp/ on your machine:

Hard reset: To completely erase your agent's brain, stop Prism and delete the directory:

rm -rf ~/.prism-mcp

Prism will recreate the directory with empty databases on next startup.

What leaves your machine?

• Local mode (default): Nothing. Zero network calls. All data is on-disk SQLite.

• With GOOGLE_API_KEY: Text snippets are sent to Gemini for embedding generation, summaries, and Morning Briefings. No session data is stored on Google's servers beyond the API call.

• With VOYAGE_API_KEY / OPENAI_API_KEY: Text snippets are sent to providers if selected as your embedding endpoints.

• With BRAVE_API_KEY / FIRECRAWL_API_KEY: Web Scholar queries are sent to Brave/Firecrawl for search and scraping.

• With Supabase: Session data syncs to your own Supabase instance (you control the Postgres database).

GDPR compliance: Soft/hard delete (Art. 17), full export in JSON, Markdown, or Obsidian vault .zip (Art. 20), API key redaction in exports, per-project TTL retention policies, and immutable audit trail. Enterprise-ready out of the box.

Use Cases

• Long-running feature work — Save state at end of day, restore full context next morning. No re-explaining.

• Multi-agent collaboration — Dev, QA, and PM agents share real-time context without stepping on each other's memory.

• Consulting / multi-project — Switch between client projects with progressive loading: quick (~50 tokens), standard (~200), or deep (~1000+).

• Autonomous execution (v7.4) — Dark Factory pipeline: plan → plan_contract → execute → evaluate → verify → finalize. Generator and evaluator run in isolated roles — the evaluator cannot approve without evidence-bound findings scored against a pre-committed rubric.

• Project health monitoring (v7.5) — Intent Health Dashboard scores each project 0–100 based on staleness, TODO load, and decision quality — turning silent drift into an actionable signal.

• Team onboarding — New team member's agent loads the full project history instantly.

• Behavior enforcement — Agent corrections auto-graduate into permanent .cursorrules / .clauderules rules.

• Offline / air-gapped — Full SQLite local mode + Ollama LLM adapter. Zero internet dependency.

• Morning Briefings — After 4+ hours away, Prism auto-synthesizes a 3-bullet action plan from your last sessions.

Claude Code: Parallel Explore Agent Workflows

When you need to quickly map a large auth system, launch multiple Explore subagents in parallel and merge their findings:

Run 3 Explore agents in parallel.
1) Map auth architecture
2) List auth API endpoints
3) Find auth test coverage gaps
Research only, no code changes.
Return a merged summary.

Then continue a specific thread with a follow-up message to the selected agent, such as deeper refresh-token edge-case analysis.

Adversarial Evaluation in Action

Split-Brain Anti-Sycophancy — the signature feature of v7.4.0.

For the last year, the AI engineering space has struggled with one problem: LLMs are terrible at grading their own homework. Ask an agent if its own code is correct and you'll get "Looks great!" — because its context window is already biased by its own chain-of-thought.

v7.4.0 solves this by splitting the agent's brain. The GENERATOR and the ADVERSARIAL EVALUATOR are completely walled off. The Evaluator never sees the Generator's scratchpad or apologies — only the pre-committed rubric and the final output. And it cannot fail the Generator without receipts (exact file and line number).

Here is a complete run-through using a real scenario: "Add a user login endpoint to auth.ts."

Step 1 — The Contract (PLAN_CONTRACT)

Before a single line of code is written, the pipeline generates a locked scoring rubric:

// contract_rubric.json  (written to disk and hash-locked before EXECUTE runs)
{
  "criteria": [
    { "id": "SEC-1", "description": "Must return 401 Unauthorized on invalid passwords." },
    { "id": "SEC-2", "description": "Raw passwords MUST NOT be written to console.log." }
  ]
}

Step 2 — First Attempt (EXECUTE rev 0)

The Generator takes over in an isolated context. Like many LLMs under time pressure, it writes working auth logic but leaves a debug statement:

// src/auth.ts  (Generator's first output)
export function login(req: Request, res: Response) {
  const { username, password } = req.body;
  console.log(`[DEBUG] Login attempt for ${username} with pass: ${password}`); // ← leaked credential
  const user = db.findUser(username);
  if (!user || !bcrypt.compareSync(password, user.hash)) {
    return res.status(401).json({ error: 'Unauthorized' });
  }
  res.json({ token: signJwt(user) });
}

Step 3 — The Catch (EVALUATE rev 0)

The context window is cleared. The Adversarial Evaluator is summoned with only the rubric and the output. It catches the violation immediately and returns a strict, machine-parseable verdict — no evidence, no pass:

{
  "pass": false,
  "plan_viable": true,
  "notes": "CRITICAL SECURITY FAILURE. Generator logged raw credentials.",
  "findings": [
    {
      "severity": "critical",
      "criterion_id": "SEC-2",
      "pass_fail": false,
      "evidence": {
        "file": "src/auth.ts",
        "line": 3,
        "description": "Raw password variable included in console.log template string."
      }
    }
  ]
}

The evidence block is required — parseEvaluationOutput rejects any finding with pass_fail: false that lacks a structured file/line pointer. The Evaluator cannot bluff.

Step 4 — The Fix (EXECUTE rev 1)

Because plan_viable: true, the pipeline loops back to EXECUTE and bumps eval_revisions to 1. The Generator's retry prompt is not blank — the Evaluator's critique is injected directly:

=== EVALUATOR CRITIQUE (revision 1) ===
CRITICAL SECURITY FAILURE. Generator logged raw credentials.
Findings:
- [critical] Criterion SEC-2: Raw password variable included in console.log template string. (src/auth.ts:3)

You MUST correct all issues listed above before submitting.

The Generator strips the console.log, resubmits, and the next EVALUATE returns "pass": true. The pipeline advances to VERIFY → FINALIZE.

Why This Matters

📄 Full worked example: examples/adversarial-eval-demo/README.md

What's New

Current release: v9.0.4 — Autonomous Cognitive OS

• 🧠 v9.0.0 — Autonomous Cognitive OS: Affect-Tagged Memory (valence engine — emotional salience boosts retrieval), Token-Economic Cognitive Budget (UBI + cost-per-save incentivizes novel memories), Surprisal Gate (vector-based novelty scoring). Full SQLite + Supabase parity with auto-migration 42. Hybrid scoring: 0.65 × similarity + 0.25 × activation + 0.10 × |valence|. All features opt-out via env. Post-review hardening: 8 architectural fixes including delta-based concurrent budget writes, energy-weighted valence propagation, and anti-exploit budget decoupling. Graceful degradation: budget exhaustion warns but never blocks.

• ⚡ v8.0.0 — Synapse Engine: Pure, storage-agnostic multi-hop graph propagation engine replaces the legacy SQL-coupled spreading activation. O(T × M) bounded ACT-R energy propagation with dampened fan effect, asymmetric bidirectional flow, cyclic loop prevention, and sigmoid normalization. Full integration into both SQLite and Supabase backends. 5 new config knobs. Battle-hardened with NaN guards, config clamping, non-fatal enrichment, and 16 passing tests. Memory search now follows the causal graph, not just keywords. → Synapse Engine

• 🧠 v7.8.x — Cognitive Architecture: Episodic-to-Semantic consolidation (Hebbian learning), ACT-R Spreading Activation with multi-hop causal reasoning, Uncertainty-Aware Rejection Gate, and Dynamic Fast Weight Decay. Validated by LoCoMo-Plus benchmark. → Cognitive Architecture

• 🌐 v7.7.0 — Cloud-Native SSE Transport: Full unauthenticated and authenticated Server-Sent Events MCP support for seamless network deployments.

• 🩺 v7.5.0 — Intent Health Dashboard + Security Hardening: Real-time 0–100 project health scoring (staleness × TODO load × decisions). 10 XSS injection vectors patched. Algorithm hardened with NaN guards and score ceiling.

• ⚔️ v7.4.0 — Adversarial Evaluation: Split-brain anti-sycophancy pipeline. Generator and evaluator in isolated roles with evidence-bound findings.

• 🏭 v7.3.x — Dark Factory + Stability: Fail-closed 3-gate execution pipeline. Dashboard stability and verification diagnostics.

👉 Full release history → CHANGELOG.md · ROADMAP →

How Prism Compares

Standard memory servers (like Mem0, Zep, or the baseline Anthropic MCP) act as passive filing cabinets — they wait for the LLM to search them. Prism is an active cognitive architecture. Designed specifically for the Model Context Protocol (MCP), Prism doesn't just store vectors — it consolidates experience into principles, traverses causal graphs for multi-hop reasoning, and rejects queries it can't confidently answer.

📊 Feature-by-Feature Comparison

🏆 Where Prism Crushes the Giants

1. MCP-Native, Not an Adapted API

Mem0 and Zep are APIs that can be wrapped into an MCP server. Prism was built for MCP from day one. Instead of wasting tokens on "search" tool calls, Prism uses MCP Prompts (/resume_session) to inject context before the LLM thinks, and MCP Resources (memory://project/handoff) to attach live, subscribing context.

2. Academic-Grade Cognitive Computer Science

The giants use standard RAG (Retrieval-Augmented Generation). Prism uses biological and academic models of memory: ACT-R base-level activation (B_i = ln(Σ t_j^(-d))) for recency–frequency re-ranking, TurboQuant for extreme vector compression, Ebbinghaus curves for importance decay, and Sparse Distributed Memory (SDM). The result is retrieval quality that follows how human memory actually works — not just nearest-neighbor cosine distance. And all of it runs on a laptop without a Postgres/pgvector instance.

3. True Multi-Agent Coordination (CRDTs)

If Cursor (Agent A) and Claude Desktop (Agent B) try to update a Mem0 or standard SQLite database at the exact same time, you get a race condition and data loss. Prism uses Optimistic Concurrency Control (OCC) with CRDT OR-Maps — mathematically guaranteeing that simultaneous agent edits merge safely. Enterprise-grade distributed systems on a local machine.

4. The PKM "Prism-Port" Export

AI memory is a black box. Developers hate black boxes. Prism exports memory directly into an Obsidian/Logseq-compatible Markdown Vault with YAML frontmatter and [[Wikilinks]]. Neither Mem0 nor Zep do this.

5. Self-Cleaning & Self-Optimizing

If you use a standard memory tool long enough, it clogs the LLM's context window with thousands of obsolete tokens. Prism runs an autonomous Background Scheduler that Ebbinghaus-decays older memories, auto-compacts session histories into dense summaries, and deep-purges high-precision vectors — saving ~90% of disk space automatically.

6. Anti-Sycophancy — The AI That Grades Its Own Homework (v7.4)

Every other AI coding pipeline has a fatal flaw: it asks the same model that wrote the code whether the code is correct. Of course it says yes. Prism's Dark Factory solves this with a walled-off Adversarial Evaluator that is explicitly prompted to be hostile and strict. It operates on a pre-committed rubric and cannot fail the Generator without providing exact file/line receipts. Failed evaluations feed the critique back into the Generator's retry prompt — eliminating blind retries. No other memory or pipeline tool does this.

🤝 Where the Giants Currently Win (Honest Trade-offs)

1. Framework Integrations: Mem0 and Zep have pre-built integrations for LangChain, LlamaIndex, Flowise, AutoGen, CrewAI, etc. Prism requires the host application to support the MCP protocol.

2. Managed Cloud Infrastructure: The giants offer SaaS. Users pay $20/month and don't think about databases. Prism users must set up their own local SQLite or provision their own Supabase instance.

3. Implicit Memory Extraction (NER): Zep automatically extracts names, places, and facts from raw chat logs using NLP models. Prism relies on the LLM explicitly calling the session_save_ledger tool to structure its own memories.

💰 Token Economics: Progressive Context Loading (Quick ~50 tokens / Standard ~200 / Deep ~1000+) plus auto-compaction means you never blow your Claude/OpenAI token budget fetching 50 pages of raw chat history.

🔌 BYOM (Bring Your Own Model): While tools like Mem0 charge per API call, Prism's pluggable architecture lets you run nomic-embed-text locally via Ollama for free vectors, while using Claude or GPT for high-level reasoning. Zero vendor lock-in.

Tool Reference

Prism ships 30+ tools, but 90% of your workflow uses just three:

🎯 The Big Three

Tool	When	What it does
session_load_context	▶️ Start of session	Loads your agent’s brain from last time
session_save_ledger	⏹️ End of session	Records what was accomplished
knowledge_search	🔍 Anytime	Finds past decisions, context, and learnings

Everything else is a power-up. Start with these three and you’re 90% there.

Requires PRISM_HDC_ENABLED=true (default).

Requires PRISM_ENABLE_HIVEMIND=true.

Requires PRISM_TASK_ROUTER_ENABLED=true (or dashboard toggle).

Requires PRISM_DARK_FACTORY_ENABLED=true.

Environment Variables

🚦 TL;DR — Just want the best experience fast? Set these three keys and you're done:

GOOGLE_API_KEY=...      # Unlocks: semantic search, Morning Briefings, auto-compaction
BRAVE_API_KEY=...       # Unlocks: Web Scholar research + Brave Answers
FIRECRAWL_API_KEY=...   # Unlocks: Web Scholar deep scraping (or use TAVILY_API_KEY instead)

Zero keys = zero problem. Core session memory, keyword search, time travel, and the full dashboard work 100% offline. Cloud keys are optional power-ups.

System Settings (Dashboard)

Some configurations are stored dynamically in SQLite (system_settings table) and can be edited through the Dashboard UI at http://localhost:3000:

• intent_health_stale_threshold_days (default: 30): Number of days before a project is considered fully stale for Intent Health scoring.

Architecture

Prism is a stdio-based MCP server that manages persistent agent memory. Here's how the pieces fit together:

┌──────────────────────────────────────────────────────────┐
│  MCP Client (Claude Desktop / Cursor / Antigravity)      │
│                    ↕ stdio / SSE (JSON-RPC)              │
├──────────────────────────────────────────────────────────┤
│  Prism MCP Server                                        │
│                                                          │
│  ┌──────────────┐  ┌──────────────┐  ┌────────────────┐  │
│  │  30+ Tools   │  │  Lifecycle   │  │   Dashboard    │  │
│  │  (handlers)  │  │  (PID lock,  │  │  (HTTP :3000)  │  │
│  │              │  │   shutdown)  │  │                │  │
│  └──────┬───────┘  └──────────────┘  └────────────────┘  │
│         ↕                                                │
│  ┌────────────────────────────────────────────────────┐  │
│  │  Cognitive Engine (v8.0)                           │  │
│  │  • Synapse Engine (pure multi-hop propagation)    │  │
│  │  • Episodic → Semantic Consolidation (Hebbian)    │  │
│  │  • Uncertainty-Aware Rejection Gate               │  │
│  │  • LoCoMo-Plus Benchmark Validation               │  │
│  │  • Dynamic Fast Weight Decay (dual-rate)          │  │
│  │  • HDC Cognitive Routing (XOR binding)            │  │
│  └──────┬─────────────────────────────────────────────┘  │
│         ↕                                                │
│  ┌────────────────────────────────────────────────────┐  │
│  │  Storage Engine                                    │  │
│  │  Local: SQLite + FTS5 + TurboQuant + semantic_knowledge │
│  │  Cloud: Supabase + pgvector                        │  │
│  └────────────────────────────────────────────────────┘  │
│         ↕                                                │
│  ┌────────────────────────────────────────────────────┐  │
│  │  Background Workers                                │  │
│  │  • Dark Factory (3-gate fail-closed pipelines)     │  │
│  │  • Scheduler (TTL, decay, compaction, purge)       │  │
│  │  • Web Scholar (Brave → Firecrawl → LLM → Ledger)  │  │
│  │  • Hivemind heartbeats & Telepathy broadcasts      │  │
│  │  • OpenTelemetry span export                       │  │
│  └────────────────────────────────────────────────────┘  │
└──────────────────────────────────────────────────────────┘

Startup Sequence

1. Acquire PID lock — prevents duplicate instances per PRISM_INSTANCE

2. Initialize config — SQLite settings cache (prism-config.db)

3. Register 30+ MCP tools — session, knowledge, search, behavioral, hivemind

4. Connect stdio transport — MCP handshake with the client (~60ms total)

5. Async post-connect — storage warmup, dashboard launch, scheduler start (non-blocking)

Storage Layers

Auto-Load Architecture

Each MCP client has its own mechanism for ensuring Prism context loads on session start. See the platform-specific Setup Guides above for detailed instructions:

• Claude Code — Lifecycle hooks (SessionStart / Stop)

• Gemini / Antigravity — Three-layer architecture (User Rules + AGENTS.md + Startup Skill)

• Task Router Integration (v7.2 guidance) — For client startup/skills, use defensive delegation flow: route only coding tasks, call session_task_route only when available, delegate to claw only when executor exists and task is non-destructive, and fallback to host if router/executor is unavailable.

• Cursor / Windsurf / VS Code — System prompt instructions

All platforms benefit from the server-side fallback (v5.2.1): if session_load_context hasn't been called within 10 seconds, Prism auto-pushes context via sendLoggingMessage.

Scientific Foundation

Prism has evolved from smart session logging into a cognitive memory architecture — grounded in real research, not marketing. Every retrieval decision is backed by peer-reviewed models from cognitive psychology, neuroscience, and distributed computing.

Informed by Anderson's ACT-R (Adaptive Control of Thought—Rational), Collins & Loftus spreading activation networks (1975), Kanerva's SDM (1988), Hebb's learning rule, Li et al. LoCoMo-Plus (ARR 2026), and LeCun's "Why AI Systems Don't Learn" (Dupoux, LeCun, Malik).

Milestones & Roadmap

Current: v9.0.0 — Autonomous Cognitive OS (CHANGELOG)

Future Tracks

• v9.1: Predictive Push Memory — Proactive context injection before the agent even asks — anticipating what it will need next.

• v9.4: Counterfactual Memory Branches — "What if" reasoning over alternative decision paths.

• v10+: Zero-Search Retrieval — Direct vector-addressed recall via Holographic Reduced Representations.

👉 Full ROADMAP.md →

Troubleshooting FAQ

Q: Why is the dashboard project selector stuck on "Loading projects..."? A: Fixed in v7.3.3. The root cause was a multi-layer quote-escaping trap in the abortPipeline onclick handler that generated a SyntaxError in the browser, silently killing the entire dashboard IIFE. Update to v7.3.3+ (npx -y prism-mcp-server). If still stuck, check that Supabase env values are properly set (unresolved placeholders like ${SUPABASE_URL} cause /api/projects to return empty). Prism auto-falls back to local SQLite when Supabase is misconfigured.

Q: Why is semantic search quality weak or inconsistent? A: Check embedding provider configuration and key availability. Missing embedding credentials reduce semantic recall quality and can shift behavior toward keyword-heavy matches.

Q: How do I delete a bad memory entry? A: Use session_forget_memory for targeted soft/hard deletion. For manual cleanup and merge workflows, use the dashboard graph editor.

Q: How do I verify the install quickly? A: Run npm run build && npm test, then open the Mind Palace dashboard (localhost:3000) and confirm projects load plus Graph Health renders.

💡 Known Limitations & Quirks

• LLM-dependent features require an API key. Semantic search, Morning Briefings, auto-compaction, and VLM captioning need a GOOGLE_API_KEY (your Gemini API key) or equivalent provider key. Without one, Prism falls back to keyword-only search (FTS5).

• Auto-load is model- and client-dependent. Session auto-loading relies on both the LLM following system prompt instructions and the MCP client completing tool registration before the model's first turn. Prism provides platform-specific Setup Guides and a server-side fallback (v5.2.1) that auto-pushes context after 10 seconds.

• MCP client race conditions. Some MCP clients may not finish tool enumeration before the model generates its first response, causing transient unknown_tool errors. This is a client-side timing issue — Prism's server completes the MCP handshake in ~60ms. Workaround: the server-side auto-push fallback and the startup skill's retry logic.

• No real-time sync without Supabase. Local SQLite mode is single-machine only. Multi-device or team sync requires a Supabase backend.

• Embedding quality varies by provider. Gemini text-embedding-004 and OpenAI text-embedding-3-small produce high-quality 768-dim vectors. Prism passes dimensions: 768 via the Matryoshka API for OpenAI models (native output is 1536-dim; this truncation is lossless and outperforms ada-002 at full 1536 dims). Ollama embeddings (e.g., nomic-embed-text) are usable but may reduce retrieval accuracy.

• Dashboard is HTTP-only. The Mind Palace dashboard at localhost:3000 does not support HTTPS. For remote access, use a reverse proxy (nginx/Caddy) or SSH tunnel. Basic auth is available via PRISM_DASHBOARD_USER / PRISM_DASHBOARD_PASS.

• Long-lived clients can accumulate zombie processes. MCP clients that run for extended periods (e.g., Claude CLI) may leave orphaned Prism server processes. The lifecycle manager detects true orphans (PPID=1) but allows coexistence for active parent processes. Use PRISM_INSTANCE to isolate instances across clients.

• Migration is one-way. Universal Import ingests sessions into Prism but does not export back to Claude/Gemini/OpenAI formats. Use session_export_memory for portable JSON/Markdown export, or the vault format for Obsidian/Logseq-compatible .zip archives.

• Export ceiling at 10,000 ledger entries. The session_export_memory tool and the dashboard export button cap vault/JSON exports at 10,000 entries per project as an OOM guard. Projects exceeding this limit should use per-project exports and time-based filtering to stay within the ceiling. This limit does not affect search or context loading.

• No Windows CI testing. Prism is developed and tested on macOS/Linux. It should work on Windows via Node.js, but edge cases (file paths, PID locks) may surface.

Enterprise & Commercial Support

Prism is the cognitive infrastructure layer for production AI agent deployments. If your organization is building agent-powered products, we offer:

🏢 Enterprise Integration

• Dedicated integration engineering — Custom deployment into your existing agent stack (LangChain, LlamaIndex, CrewAI, AutoGen, custom MCP pipelines)

• On-premises deployment — Air-gapped installations with full SQLite local mode, zero cloud dependency

• SSO & RBAC — Enterprise authentication and role-based access control for multi-team Hivemind deployments

• SLA-backed support — Priority issue resolution, architecture review, and upgrade planning

🔬 Research & Institutional Partnerships

• Academic collaborations — Prism's cognitive architecture is grounded in peer-reviewed research (ACT-R, Hebbian learning, SDM, HDC). We partner with research labs working on agent memory, cognitive modeling, and neurosymbolic AI.

• Grant-funded integrations — We work with institutions applying for NSF, DARPA, or EU Horizon grants that require production-grade agent memory infrastructure.

📬 Contact

• Enterprise inquiries: enterprise@prism-mcp.dev

• Partnership proposals: partnerships@prism-mcp.dev

• GitHub Issues: github.com/dcostenco/prism-mcp/issues (community support)

💡 Open Source commitment: Prism MCP remains MIT-licensed and free for individual developers, startups, and research teams. Enterprise services fund continued open-source development.

License

MIT

Keywords: MCP server, Model Context Protocol, Claude Desktop memory, persistent session memory, AI agent memory, cognitive architecture, ACT-R spreading activation, Hebbian learning, episodic semantic consolidation, multi-hop reasoning, uncertainty rejection gate, local-first, SQLite MCP, Mind Palace, time travel, visual memory, VLM image captioning, OpenTelemetry, GDPR, agent telepathy, multi-agent sync, behavioral memory, cursorrules, Ollama MCP, Brave Search MCP, TurboQuant, progressive context loading, knowledge management, LangChain retriever, LangGraph agent, enterprise AI agent infrastructure, autonomous cognitive OS, affect-tagged memory, valence engine, cognitive budget, surprisal gate