### Setup and Run LongMemEval Benchmark (Python) Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/README.md Install dependencies and run the LongMemEval benchmark using Python. This command assumes data has been downloaded from HuggingFace. ```bash cd longmemeval pip install -r requirements.txt # Download data from HuggingFace (see README) # Then run: python run.py --data data/longmemeval_oracle.json ``` -------------------------------- ### Install and Initialize Hippo CLI Source: https://github.com/kitfunso/hippo-memory/blob/master/extensions/claude-code-plugin/README.md Install the hippo CLI globally and then initialize it within your project directory. ```bash npm install -g hippo-memory ``` ```bash cd your-project hippo init ``` -------------------------------- ### Install Hippo CLI and Initialize Source: https://github.com/kitfunso/hippo-memory/blob/master/extensions/pi-extension/README.md Install the hippo CLI globally and initialize hippo in your project. You can also scan all repositories for initialization. ```bash # Install hippo CLI npm install -g hippo-memory # Init hippo in your project (or scan all repos) hippo init # or: hippo init --scan ~ ``` -------------------------------- ### Install Dependencies Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/longmemeval/README.md Install the required Python packages for the benchmark. ```bash pip install -r requirements.txt ``` -------------------------------- ### Install Dependencies and Build UI Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-16-brain-observatory-v025.md Commands to install npm dependencies and build the UI project. Assumes you are in the 'ui' directory. ```bash cd C:/Users/skf_s/hippo/ui && npm install ``` ```bash cd C:/Users/skf_s/hippo/ui && npm run build ``` -------------------------------- ### Server Bearer Lockdown Test Setup Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-29-e1.3-slack-ingestion.md Sets up the test environment by initializing the store, starting the server with authentication enabled, and defining routes to test. ```typescript import { describe, it, expect, beforeEach, afterEach } from 'vitest'; import { mkdtempSync, rmSync } from 'fs'; import { tmpdir } from 'os'; import { join } from 'path'; import { initStore } from '../src/store.js'; import { serve } from '../src/server.js'; describe('server Bearer lockdown', () => { let root: string; let handle: { stop: () => Promise; port: number }; beforeEach(async () => { root = mkdtempSync(join(tmpdir(), 'hippo-lockdown-')); initStore(root); process.env.HIPPO_REQUIRE_AUTH = '1'; handle = await serve({ hippoRoot: root, host: '127.0.0.1', port: 0 }); }); afterEach(async () => { delete process.env.HIPPO_REQUIRE_AUTH; await handle.stop(); rmSync(root, { recursive: true, force: true }); }); const v1Routes: Array<{ method: string; path: string; body?: string }> = [ { method: 'POST', path: '/v1/memories', body: '{"content":"x"}' }, { method: 'GET', path: '/v1/memories?q=x' }, { method: 'POST', path: '/v1/auth/keys', body: '{}' }, { method: 'GET', path: '/v1/auth/keys' }, { method: 'GET', path: '/v1/audit' }, ]; it('all /v1/* routes except slack require Bearer', async () => { for (const r of v1Routes) { const res = await fetch(`http://127.0.0.1:${handle.port}${r.path}`, { method: r.method, headers: { 'content-type': 'application/json' }, body: r.body, }); expect(res.status, `${r.method} ${r.path}`).toBe(401); } }); }); ``` -------------------------------- ### Hippo Memory Installation and Initialization Source: https://github.com/kitfunso/hippo-memory/blob/master/extensions/mcp/README.md Install the hippo-memory package globally using npm and then initialize it in your project directory. ```bash npm install -g hippo-memory cd your-project && hippo init ``` -------------------------------- ### Copy Pi Extension for Project-Level Installation Source: https://github.com/kitfunso/hippo-memory/blob/master/extensions/pi-extension/README.md Copy the pi-extension directory to the project's .pi/extensions directory for project-level installation. ```bash cp -r extensions/pi-extension .pi/extensions/hippo-memory ``` -------------------------------- ### Install Dependencies Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-29-b3-dlpfc-depth.md Install project dependencies using npm. ```bash npm install ``` -------------------------------- ### Install and Initialize Hippo Memory CLI Source: https://context7.com/kitfunso/hippo-memory/llms.txt Install the hippo-memory package globally and initialize a project store. The `--scan` option can discover all git repositories under a given path. Use `--no-hooks` and `--no-schedule` to initialize without modifying agent configuration files. ```bash # Install npm install -g hippo-memory # Initialize a single project (auto-installs agent hooks into CLAUDE.md / AGENTS.md / .cursorrules) cd my-project && hippo init # Initialize all projects under home directory at once hippo init --scan ~ # Initialize without touching agent config files hippo init --no-hooks --no-schedule # Check what was created hippo status # Memories: 0 (0 pinned, 0 errors) # Avg strength: 0 # At risk (<0.1): 0 # Open conflicts: 0 # Half-life default: 7d ``` -------------------------------- ### Install Hippo Memory Plugin Source: https://github.com/kitfunso/hippo-memory/blob/master/extensions/claude-code-plugin/README.md Use these commands to install the Hippo Memory plugin from the marketplace or locally for development. ```bash /plugin marketplace add kitfunso/hippo-memory /plugin install hippo-memory ``` ```bash claude --plugin-dir ./extensions/claude-code-plugin ``` ```bash cp -r extensions/claude-code-plugin ~/.claude/plugins/hippo-memory ``` -------------------------------- ### Initialize and Start Rendering (JavaScript) Source: https://github.com/kitfunso/hippo-memory/blob/master/mockups/medical-imaging.html Initializes the visualization by setting up event listeners for window resize and starting the animation loop. Ensures the canvas is responsive to display size changes. ```javascript window.addEventListener('resize', resize); resize(); requestAnimationFrame(render); })(); ``` -------------------------------- ### Manual Installation of Hippo Memory Plugin Source: https://github.com/kitfunso/hippo-memory/blob/master/extensions/openclaw-plugin/README.md Manually copy the plugin files to the OpenClaw extensions directory if not using the plugin installer. ```bash cp -r extensions/openclaw-plugin ~/.openclaw/extensions/hippo-memory ``` -------------------------------- ### Install Hippo Memory Plugin Source: https://github.com/kitfunso/hippo-memory/blob/master/extensions/openclaw-plugin/README.md Install and enable the Hippo Memory plugin directly from npm or link a local folder for development. ```bash openclaw plugins install hippo-memory openclaw plugins enable hippo-memory ``` ```bash openclaw plugins install -l ./extensions/openclaw-plugin ``` -------------------------------- ### JavaScript for Celestial Scene Setup Source: https://github.com/kitfunso/hippo-memory/blob/master/mockups/bg-celestial-star.html Initializes Three.js, scene, camera, renderer, and OrbitControls. Sets up the background color and fog. ```javascript { "imports": { "three": "https://unpkg.com/three@0.170.0/build/three.module.js", "three/addons/": "https://unpkg.com/three@0.170.0/examples/jsm/" }} import * as THREE from "three"; import { OrbitControls } from "three/addons/controls/OrbitControls.js"; const PAL = { teal: new THREE.Color("#4a8ca3"), sage: new THREE.Color("#5a8f6b"), violet: new THREE.Color("#7c6caf"), }; const LAYER_Y = { buffer: 14, episodic: 7, semantic: 0 }; function mulberry(a) { return function() { a |= 0; a = a + 0x6D2B79F5 | 0; let t = Math.imul(a ^ a >>> 15, 1 | a); t = t + Math.imul(t ^ t >>> 7, 61 | t) ^ t; return ((t ^ t >>> 14) >>> 0) / 4294967296; }; } const rand = mulberry(314); const MEMORIES = []; const layers = ["buffer","episodic","semantic"]; for (let i = 0; i < 28; i++) { MEMORIES.push({ id: `m${i}`, layer: layers[i % 3], strength: 0.3 + rand() * 0.7, text: `word${i} phrase${i%5}` }); } const positions = MEMORIES.map((m, i) => { const angle = i * 2.3998; const r = 6 + Math.sqrt(i) * 3.4; return new THREE.Vector3(Math.cos(angle) * r, LAYER_Y[m.layer], Math.sin(angle) * r); }); const mount = document.getElementById("scene"); const scene = new THREE.Scene(); scene.background = new THREE.Color("#f0e7d2"); scene.fog = new THREE.Fog(0xf0e7d2, 55, 130); const camera = new THREE.PerspectiveCamera(42, window.innerWidth / window.innerHeight, 0.1, 400); camera.position.set(26, 22, 38); const renderer = new THREE.WebGLRenderer({ antialias: true }); renderer.setPixelRatio(Math.min(2, window.devicePixelRatio)); renderer.setSize(window.innerWidth, window.innerHeight); mount.appendChild(renderer.domElement); const controls = new OrbitControls(camera, renderer.domElement); controls.enableDamping = true; controls.target.set(0, 6, 0); // ------- Celestial chart floor: concentric rings + radial lines + zodiac ticks function makeCelestialTexture() { const size = 1024; const c = document.createElement("canvas"); c.width = c.height = size; const ctx = c.getContext("2d"); ctx.fillStyle = "#f0e7d2"; ctx.fillRect(0,0,size,size); // grain const img = ctx.getImageData(0,0,size,size); for (let i = 0; i < img.data.length; i += 4) { const n = (Math.random() - 0.5) * 20; img.data[i] = Math.max(0, Math.min(255, img.data[i] + n)); img.data[i+1] = Math.max(0, Math.min(255, img.data[i+1] + n - 2)); img.data[i+2] = Math.max(0, Math.min(255, img.data[i+2] + n - 10)); } ctx.putImageData(img, 0, 0); const cx = size/2, cy = size/2; // concentric circles ctx.strokeStyle = "rgba(31,25,18,0.35)"; ctx.lineWidth = 1.4; for (let r = 60; r < size/2; r += 70) { ctx.beginPath(); ctx.arc(cx, cy, r, 0, Math.PI*2); ctx.stroke(); } // faint in-between ctx.strokeStyle = "rgba(31,25,18,0.14)"; ctx.lineWidth = 1; for (let r = 95; r < size/2; r += 70) { ctx.beginPath(); ctx.arc(cx, cy, r, 0, Math.PI*2); ctx.stroke(); } // radial lines (spokes) — 16 major ctx.strokeStyle = "rgba(31,25,18,0.28)"; ctx.lineWidth = 1.1; for (let i = 0; i < 16; i++) { const a = (i / 16) * Math.PI * 2; ctx.beginPath(); ctx.moveTo(cx + Math.cos(a) * 40, cy + Math.sin(a) * 40); ctx.lineTo(cx + Math.cos(a) * (size/2 - 30), cy + Math.sin(a) * (size/2 - 30)); ctx.stroke(); } // minor ticks — 64 ctx.strokeStyle = "rgba(31,25,18,0.22)"; ctx.lineWidth = 1; for (let i = 0; i < 64; i++) { const a = (i / 64) * Math.PI * 2; ctx.beginPath(); ctx.moveTo(cx + Math.cos(a) * (size/2 - 50), cy + Math.sin(a) * (size/2 - 50)); ctx.lineTo(cx + Math.cos(a) * (size/2 - 32), cy + Math.sin(a) * (size/2 - 32)); ctx.stroke(); } // compass rose at center ctx.save(); ctx.translate(cx, cy); ctx.fillStyle = "rgba(31,25,18,0.55)"; ctx.strokeStyle = "rgba(31,25,18,0.75)"; ctx.lineWidth = 1.4; for (let i = 0; i < 4; i++) { ctx.rotate(Math.PI/2); ctx.beginPath(); ctx.moveTo(0, 0); ctx.lineTo(12, -34); ctx.lineTo(0, -54); ctx.lineTo(-12, -34); ctx.closePath(); ctx.fill(); ctx.stroke(); } ctx.restore(); // cardinal letters ctx.fillStyle = "rgba(31,25,18,0.7)"; ctx.font = "bold 22px Georgia"; ctx.textAlign = "center"; ctx.textBaseline = "middle"; ctx.fillText("N", cx, cy - size/2 + 40); ctx.fillText("S", cx, cy + size/2 - 40); ctx.fillText("E", cx + size/2 - 40, cy); ctx.fillText("W", cx - size/2 + 40, cy); return c; } ``` -------------------------------- ### MCP Server Configuration Source: https://context7.com/kitfunso/hippo-memory/llms.txt Example configuration file (`mcp.json`) for setting up the Hippo MCP server to run as a command. ```json // .cursor/mcp.json or claude_desktop_config.json { "mcpServers": { "hippo-memory": { "command": "hippo", "args": ["mcp"] } } } ``` -------------------------------- ### Start Hippo Memory API Server Source: https://context7.com/kitfunso/hippo-memory/llms.txt Starts an HTTP daemon that exposes the full memory API. It supports bearer-token authentication and audit logging. By default, it serves on loopback only. ```bash hippo serve --port 6789 ``` -------------------------------- ### MCP Server Configuration Example Source: https://github.com/kitfunso/hippo-memory/blob/master/README.md Example JSON configuration for an MCP client to connect to the Hippo Memory MCP server. This should be added to your MCP client's configuration file. ```json { "mcpServers": { "hippo-memory": { "command": "hippo", "args": ["mcp"] } } } ``` -------------------------------- ### Initialization and Game Start Source: https://github.com/kitfunso/hippo-memory/blob/master/mockups/art-hippocampus-anime.html Initializes the canvas, sets up event listeners, pushes an initial log message, renders the timeline, and starts the main game loop using requestAnimationFrame. ```javascript resize(); pushLog("monitor online hippo-chan greets you"); renderTL(); requestAnimationFrame(tick); })(); ``` -------------------------------- ### Start MCP Server Source: https://github.com/kitfunso/hippo-memory/blob/master/README.md Starts the Hippo Memory MCP server over stdio. This command is used for integrating with MCP-compatible clients like Cursor, Windsurf, Cline, and Claude Desktop. ```bash hippo mcp # starts MCP server over stdio ``` -------------------------------- ### Install and Initialize Hippo Memory Source: https://github.com/kitfunso/hippo-memory/blob/master/README.md Installs the hippo-memory package globally and initializes memory for all git repositories on the machine. This command scans the home directory for git repositories and sets up a .hippo/ store in each, seeding it with recent commit history. ```bash npm install -g hippo-memory && hippo init --scan ~ ``` -------------------------------- ### Example Hook Block in CLAUDE.md Source: https://context7.com/kitfunso/hippo-memory/llms.txt This is an example of the hook block that is added to the `CLAUDE.md` file when the `claude-code` hook is installed. ```markdown ``` -------------------------------- ### Initialize UI Directory Structure Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-16-brain-observatory-v025.md Creates the necessary directories for the UI application. Ensure you are in the project's root directory before running. ```bash cd C:/Users/skf_s/hippo mkdir -p ui/src/api ui/src/views ui/src/components ``` -------------------------------- ### GitHub Actions Workflow Configuration Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-05-05-provenance-ci-gate.md Defines the CI workflow including checkout, Node.js setup, dependency installation, build, testing, and the provenance gate step. Sets a timeout for the job. ```yaml permissions: contents: read jobs: test: runs-on: ubuntu-latest # codex P1: 15min was tight on first-run npm ci with @xenova/transformers # optionalDependency (large native install). 25 gives headroom on a cold cache. timeout-minutes: 25 steps: - uses: actions/checkout@v4 - uses: actions/setup-node@v4 with: node-version: '22.5.0' cache: npm - run: npm ci - run: npm run build - name: Vitest run: npx vitest run - name: Provenance gate run: node scripts/ci-seed-provenance.mjs - name: Upload provenance JSON if: always() uses: actions/upload-artifact@v4 with: name: provenance-coverage path: provenance-coverage.json if-no-files-found: warn retention-days: 14 ``` -------------------------------- ### Hippo Recall Command with Explanation Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/working-memory-roadmap.md Example of using the `hippo recall` command with the `--why` flag to get explanations for returned items, including match reasons and source buckets. This was part of PR4. ```bash hippo recall --why --limit 5 ``` -------------------------------- ### Build and Deploy Frontend Source: https://github.com/kitfunso/hippo-memory/blob/master/tests/traps/trap-repo/README.md Use the PowerShell script to build the frontend and deploy it. ```powershell .\src\build.ps1 ``` -------------------------------- ### Manual OpenClaw Hook Configuration Source: https://github.com/kitfunso/hippo-memory/blob/master/integrations/openclaw.md Configuration for manual OpenClaw hook installation in AGENTS.md. Includes commands for session start context retrieval, error reporting, outcome logging, and learning from coding sessions. ```markdown ## Project Memory (Hippo) At the start of every session, run: hippo context --auto --budget 1500 On errors or unexpected behaviour: hippo remember "" --error On task completion: hippo outcome --good After significant coding sessions: hippo learn --git ``` -------------------------------- ### Install Dependencies Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/locomo/README.md Installs necessary Python packages. Requires the hippo CLI to be installed globally. ```bash pip install -r requirements.txt # Requires: hippo CLI installed globally (hippo --version => 0.31.0) # Judge: uses `claude -p` CLI (no ANTHROPIC_API_KEY required) ``` -------------------------------- ### Initialize and Handle Window Resize Source: https://github.com/kitfunso/hippo-memory/blob/master/mockups/hybrid-v4.html Sets up the renderer and camera aspect ratio, and adds an event listener to update them on window resize. ```javascript function resize() { const r = container.getBoundingClientRect(); renderer.setSize(r.width, r.height); camera.aspect = r.width / Math.max(r.height, 1); camera.updateProjectionMatrix(); } resize(); window.addEventListener('resize', resize); ``` -------------------------------- ### Install Codex Integration Source: https://github.com/kitfunso/hippo-memory/blob/master/integrations/codex.md Run this command to manually install or repair the Codex integration if Hippo was installed before Codex. ```bash hippo hook install codex ``` -------------------------------- ### Verify Full Build Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-16-brain-observatory-v025.md Command to build the UI project. This command should be run from the 'ui' directory and is expected to create a 'dist-ui/' folder at the project root. ```bash cd C:/Users/skf_s/hippo/ui && npm run build ``` -------------------------------- ### Manually Install OpenClaw Hook Source: https://github.com/kitfunso/hippo-memory/blob/master/integrations/openclaw.md Install the OpenClaw hook manually for lightweight prompt injection without installing the native plugin. This appends a hippo block to AGENTS.md. ```bash hippo hook install openclaw ``` -------------------------------- ### Initialize Hippo Memory Project Source: https://github.com/kitfunso/hippo-memory/blob/master/PLAN.md Use this command to set up the necessary configuration and directory structure for a new Hippo Memory project. ```bash hippo init ``` -------------------------------- ### Initialize Project Context with Hippo Source: https://github.com/kitfunso/hippo-memory/blob/master/AGENTS.md Use this command at the start of each task to set the project context and budget. Read the output before writing any code. ```bash hippo context --auto --budget 1500 ``` -------------------------------- ### Run the RSI Demo Source: https://github.com/kitfunso/hippo-memory/blob/master/examples/rsi-demo/README.md Execute the main agent script to start the demonstration. The output shows the first-attempt success rate over different task windows and the calculated learning gap. ```bash node agent.mjs ``` -------------------------------- ### Outcomes Action Example Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/micro/README.md Example of an 'outcomes' action, applying positive or negative outcomes to a remembered item. ```json {"type": "outcomes", "remember_index": 0, "good": 3, "bad": 0} ``` -------------------------------- ### Run Build and Test Suite Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-22-bi-temporal.md Execute the build process and run the full test suite to verify the migration and implementation. This command should be run after implementing the migration steps. ```bash npm run build && npx vitest run ``` -------------------------------- ### Commit UserPromptSubmit Hook Installation Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-21-pinned-reinject.md Stages and commits the changes related to installing the UserPromptSubmit hook for pinned-rule re-injection. ```bash git add src/hooks.ts tests/pinned-hook-install.test.ts git commit -m "feat(hooks): install UserPromptSubmit entry for pinned-rule re-inject" ``` -------------------------------- ### 3D Scene Setup and Animation Loop Source: https://github.com/kitfunso/hippo-memory/blob/master/mockups/bg-nebula-bio.html Sets up a 3D scene with animated orbs and pollen particles. Includes a main animation loop that updates object scales, materials, and positions, then re-renders the scene. Handles window resizing to maintain aspect ratio. ```javascript const edges = new THREE.LineSegments( new THREE.BufferGeometry().setFromPoints(ep), new THREE.LineBasicMaterial({ color: "#2a241e", transparent: true, opacity: 0.25 }) ); scene.add(edges); const clock = new THREE.Clock(); function frame() { const t = clock.getElapsedTime(); for (const { inner, outer } of orbs) { const p1 = 0.10 * Math.sin(t * 0.9 + inner.userData.phase); const p2 = 0.16 * Math.sin(t * 0.9 + outer.userData.phase); inner.scale.setScalar(inner.userData.base * (1 + p1)); outer.scale.setScalar(outer.userData.base * (1 + p2)); outer.material.opacity = 0.18 + 0.12 * Math.sin(t * 0.9 + outer.userData.phase); } // pollen drift + gentle sway, wrap when out of bounds const pArr = pollenGeom.attributes.position.array; for (let i = 0; i < POLLEN_COUNT; i++) { const k = i * 3; const sway = 0.03 * Math.sin(t * 0.7 + pollenPhase[i]); pArr[k] += pollenVel[k] + sway * 0.02; pArr[k+1] += pollenVel[k+1]; pArr[k+2] += pollenVel[k+2] + sway * 0.01; if (pArr[k+1] > 32) { pArr[k] = (rand() - 0.5) * pollenSpread * 2; pArr[k+1] = -3; pArr[k+2] = (rand() - 0.5) * pollenSpread * 2; } } pollenGeom.attributes.position.needsUpdate = true; controls.update(); renderer.render(scene, camera); requestAnimationFrame(frame); } frame(); window.addEventListener("resize", () => { camera.aspect = window.innerWidth / window.innerHeight; camera.updateProjectionMatrix(); renderer.setSize(window.innerWidth, window.innerHeight); }); ``` -------------------------------- ### Run Sequential Learning Benchmark Source: https://github.com/kitfunso/hippo-memory/blob/master/README.md Command to execute the sequential learning benchmark using all available adapters. Navigate to the benchmarks/sequential-learning directory first. ```bash cd benchmarks/sequential-learning node run.mjs --adapter all ``` -------------------------------- ### Recall Pre-Action Example Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/micro/README.md Example of a 'recall' pre-action for a query, used to bump the retrieval count of a specific memory. ```json {"type": "recall", "query": "marker-pineal-1", "remember_index": 0, "times": 3} ``` -------------------------------- ### Supersede Action Example Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/micro/README.md Example of a 'supersede' action within a fixture, updating the content of a specific remembered item. ```json {"type": "supersede", "remember_index": 0, "new_content": "Bob switched to oat milk flat white"} ``` -------------------------------- ### Build and Test Commands Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-16-brain-observatory-v025.md Commands to build the project, run tests, and start the dashboard locally. Ensure you are in the project root directory. ```bash cd C:/Users/skf_s/hippo npm run build:all npm run test node bin/hippo.js dashboard # Open http://localhost:3333 # Verify: particles render, hover tooltips work, search filters, detail panel opens, conflicts show ``` -------------------------------- ### Initialize Benchmark Runner Script Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-21-model-aware-hippo.md Sets up the main script for running model profile benchmarks. Imports necessary modules and parses command-line flags for models, profiles, and output destinations. It reads the benchmark corpus from a JSON file. ```javascript #!/usr/bin/env node import * as fs from 'node:fs'; import { execFileSync } from 'node:child_process'; import { judge } from './model-profile-judge.mjs'; const corpus = JSON.parse(fs.readFileSync('evals/model-profile-bench.json', 'utf8')); const TARGET_MODELS = flag('--models', 'claude-opus-4-6,claude-opus-4-7').split(','); const PROFILES = flag('--profiles', 'default').split(','); const OUT = flag('--out', 'evals/baseline-profile-bench.json'); // No API key check — uses claude CLI subscription ``` -------------------------------- ### Install Hippo Memory Source: https://github.com/kitfunso/hippo-memory/blob/master/README.md Installs the hippo-memory package globally using npm. This is the first step before initializing memory for projects. ```bash npm install -g hippo-memory ``` -------------------------------- ### Run Sequential Learning Benchmark Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/sequential-learning/README.md Execute the benchmark script. Options include running all conditions, a specific adapter, or customizing the output directory. ```bash # Run all conditions node benchmarks/sequential-learning/run.mjs # Run a specific adapter node benchmarks/sequential-learning/run.mjs --adapter hippo node benchmarks/sequential-learning/run.mjs --adapter static node benchmarks/sequential-learning/run.mjs --adapter none # Custom output directory node benchmarks/sequential-learning/run.mjs --output my-results/ ``` -------------------------------- ### Start stdio MCP Server Source: https://context7.com/kitfunso/hippo-memory/llms.txt Launches the Hippo MCP server to expose tools over stdio, allowing integration with compatible agents. ```bash hippo mcp ``` -------------------------------- ### Slack Webhook POST Request Example Source: https://github.com/kitfunso/hippo-memory/blob/master/docs/plans/2026-04-29-e1.3-slack-ingestion.md Example of a POST request to the Slack events endpoint with necessary headers for signature verification. ```typescript method: 'POST', headers: { 'content-type': 'application/json', 'x-slack-request-timestamp': ts, 'x-slack-signature': sign(ts, body) }, body, }); expect(res.status).toBe(200); // DLQ row exists. const { openHippoDb, closeHippoDb } = await import('../src/db.js'); const db = openHippoDb(root); try { const row = db.prepare(`SELECT COUNT(*) as c FROM slack_dlq`).get() as { c: number }; expect(Number(row.c)).toBe(1); } finally { closeHippoDb(db); } ``` -------------------------------- ### Run Sequential Learning Benchmark (Node.js) Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/README.md Execute the sequential learning benchmark using Node.js. Ensure you have Node.js 22.5+ installed. This command runs all available adapters. ```bash cd sequential-learning node run.mjs --adapter all ``` -------------------------------- ### Goal Push Pre-Action Example Source: https://github.com/kitfunso/hippo-memory/blob/master/benchmarks/micro/README.md Example of a 'goal_push' pre-action for a query, used to set a named goal for the 'hippo recall' command. ```json {"op": "goal_push", "name": "db-rewrite", "session_id": "s-db"} ``` -------------------------------- ### Initialization and Resize Handling (JavaScript) Source: https://github.com/kitfunso/hippo-memory/blob/master/mockups/art-hippocampus-breathing.html Sets up the initial state of the animation, including resizing the canvas and starting the animation loop. It also seeds the log with initial status messages. ```javascript // ---------- init ---------- resize(); // seed a couple of log lines so it's not empty on load pushLog(`monitor online :: 12.0 bpm baseline`); renderTL(); rafId = requestAnimationFrame(tick); })(); ```