diff --git a/.env.example b/.env.example
index 84d77d88..9792c34f 100644
--- a/.env.example
+++ b/.env.example
@@ -1,144 +1,32 @@
-# XActions Configuration
-# Copy this file to .env and fill in your values
+# XActions Environment Variables
+# Copy this to .env and fill in your values
+# NEVER commit .env to git
 
-# ═══════════════════════════════════════════════════════════════════════════════
-# Server Configuration
-# ═══════════════════════════════════════════════════════════════════════════════
+# X/Twitter Session Cookie
+# Get this from: DevTools (F12) -> Application -> Cookies -> x.com -> auth_token
+XACTIONS_SESSION_COOKIE=your_auth_token_here
 
-NODE_ENV=development
-PORT=3001
-API_URL=http://localhost:3001
-API_BASE_URL=http://localhost:3001
-FRONTEND_URL=http://localhost:3000
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# x402 Payment Configuration (AI Agent Monetization)
-# ═══════════════════════════════════════════════════════════════════════════════
-
-# ⚠️  REQUIRED for production - your wallet address to receive USDC payments
-# This MUST be set to a valid Ethereum address (0x...) for x402 to work
-# Server will fail to start in production if this is not configured
-X402_PAY_TO_ADDRESS=0x4027FdaC1a5216e264A00a5928b8366aE59cE888
-
-# Network: which blockchain to use for payments
-#   eip155:8453  = Base mainnet (REAL money - use for production)
-#   eip155:84532 = Base Sepolia testnet (TEST money - use for development)
-# Defaults: testnet in development, mainnet in production
-X402_NETWORK=eip155:84532
-
-# Facilitator URL for payment verification and settlement
-# Default works for most cases, only change if using custom facilitator
-X402_FACILITATOR_URL=https://x402.org/facilitator
-
-# Debug logging for x402 payments (optional)
-X402_DEBUG=false
-
-# Skip payment verification in development (NEVER use in production!)
-# Uncomment ONLY for local testing without a wallet
-# X402_SKIP_VERIFICATION=true
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# Payment Webhook Notifications (Optional - for revenue tracking)
-# ═══════════════════════════════════════════════════════════════════════════════
-
-# Custom webhook URL - receives JSON POST for all payment events
-# This is the recommended option - just point to your own endpoint
-# Payload: { event, timestamp, data: { amount, operation, payer, network, txHash } }
-X402_WEBHOOK_URL=https://your-server.com/webhooks/payments
-
-# Webhook signature secret (optional but recommended for production)
-# Signs payloads with HMAC-SHA256 for security
-# Verify: X-Webhook-Signature header = HMAC-SHA256(timestamp.payload, secret)
-# Generate with: openssl rand -hex 32
-X402_WEBHOOK_SECRET=
-
-# Optional: Discord/Slack webhooks (leave empty if not using)
-# DISCORD_WEBHOOK_URL=
-# SLACK_WEBHOOK_URL=
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# Stripe Billing (Subscription Payments)
-# ═══════════════════════════════════════════════════════════════════════════════
-
-# Get your keys from https://dashboard.stripe.com/apikeys
-STRIPE_SECRET_KEY=sk_test_...
-STRIPE_PUBLISHABLE_KEY=pk_test_...
-
-# Webhook signing secret — from Stripe Dashboard > Webhooks > Signing secret
-# Endpoint URL: https://your-domain.com/webhooks/stripe
-STRIPE_WEBHOOK_SECRET=whsec_...
-
-# Price IDs — create these in Stripe Dashboard > Products
-# Each should be a recurring monthly price
-STRIPE_PRO_PRICE_ID=price_...
-STRIPE_BUSINESS_PRICE_ID=price_...
-STRIPE_ENTERPRISE_PRICE_ID=price_...
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# MCP Server Configuration (for AI agents like Claude)
-# ═══════════════════════════════════════════════════════════════════════════════
-
-# MCP mode: local (free, uses Puppeteer) or remote (paid, uses x402 API)
+# Mode: 'local' (default, uses Puppeteer) or 'remote' (cloud API)
 XACTIONS_MODE=local
 
-# API URL for remote mode
+# Remote mode settings (only needed if XACTIONS_MODE=remote)
 XACTIONS_API_URL=https://api.xactions.app
 
-# Private key for x402 payments (remote mode only)
-# ⚠️ NEVER commit this value! Use a separate wallet for payments.
-# X402_PRIVATE_KEY=0xYourPrivateKey
-
-# X/Twitter session cookie for authentication
-XACTIONS_SESSION_COOKIE=
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# Database (PostgreSQL)
-# ═══════════════════════════════════════════════════════════════════════════════
-
-DATABASE_URL="postgresql://user:password@localhost:5432/xactions?schema=public"
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# Authentication & Security
-# ═══════════════════════════════════════════════════════════════════════════════
-
-JWT_SECRET=your-super-secret-jwt-key-change-this-in-production
-SESSION_SECRET=your-session-secret-key
-
-# Admin API key for accessing admin endpoints (payment stats, webhook management)
-# Generate a secure random string: openssl rand -hex 32
-ADMIN_API_KEY=
+# Optional: x420 micropayments (remote mode)
+X402_PRIVATE_KEY=
+X402_NETWORK=base-sepolia
 
-# CORS allowed origins (comma-separated)
-CORS_ORIGINS=http://localhost:3000,http://localhost:3001,https://xactions.app
+# Optional: AI-generated content via XActions
+OPENROUTER_API_KEY=
 
-# ═══════════════════════════════════════════════════════════════════════════════
-# Twitter/X OAuth 2.0
-# ═══════════════════════════════════════════════════════════════════════════════
+# Database (Docker Compose defaults)
+POSTGRES_PASSWORD=xactions_dev_password
+DATABASE_URL=postgresql://xactions:xactions_dev_password@localhost:5432/xactions?schema=public
 
-TWITTER_CLIENT_ID=your_twitter_client_id
-TWITTER_CLIENT_SECRET=your_twitter_client_secret
-TWITTER_BEARER_TOKEN=your_twitter_bearer_token
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# Redis (for job queue)
-# ═══════════════════════════════════════════════════════════════════════════════
-
-REDIS_HOST=localhost
-REDIS_PORT=6379
-REDIS_PASSWORD=
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# Puppeteer Configuration (for local mode)
-# ═══════════════════════════════════════════════════════════════════════════════
-
-PUPPETEER_HEADLESS=true
-PUPPETEER_NO_SANDBOX=true
-# PUPPETEER_EXECUTABLE_PATH=/usr/bin/chromium-browser
-
-# ═══════════════════════════════════════════════════════════════════════════════
-# Logging & Monitoring
-# ═══════════════════════════════════════════════════════════════════════════════
-
-LOG_LEVEL=info
-# SENTRY_DSN=https://xxx@sentry.io/xxx
+# API Server
+PORT=3001
+JWT_SECRET=change_this_to_a_random_string
+SESSION_SECRET=change_this_too
 
+# Hermes Integration
+HERMES_CONTENT_DIR=~/.hermes/cron/output
diff --git a/AGENTS.md b/AGENTS.md
index b28ef773..d19946e6 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -34,7 +34,9 @@ extension/     → Browser extension (Chrome/Edge)
 
 ## Skills
 
-31 skills in `skills/*/SKILL.md`. Read the relevant SKILL.md when a user's request matches a category.
+32 skills in `skills/*/SKILL.md`. Read the relevant SKILL.md when a user's request matches a category.
+
+- **Hermes integration** — `skills/hermes-integration/SKILL.md` — content pipeline from Hermes to X/Twitter
 
 - **Unfollow management** — mass unfollow, non-follower cleanup
 - **Analytics & insights** — engagement, hashtags, competitors, best times
@@ -104,9 +106,35 @@ Common resource hogs: `tsgo --noEmit` (~500% CPU), vitest workers (15x ~100% CPU
 - Do not reuse foreground shell sessions — stale sessions block future operations
 - If a terminal appears unresponsive, kill it and create a new one
 
+## Hermes Integration
+
+This repo is integrated with Hermes Agent for AI-powered content generation.
+
+### Content Pipeline
+
+1. Hermes generates security content (vulnerability threads, exploit analysis)
+2. All content is deslopped (no AI patterns, em dashes, hedging)
+3. Hermes sends content to the user via Telegram
+4. **User posts manually** — no automation, no API keys needed
+
+### Key Files
+
+| File | Purpose |
+|------|---------|
+| `skills/hermes-integration/SKILL.md` | Full integration guide |
+| `scripts/hermes-poster.js` | Format content for manual posting |
+| `.env.example` | Environment variable template (for future automation) |
+
+### Environment
+
+- `.env` is in `.gitignore` — never commit secrets
+- Content directory: `~/.hermes/cron/output/`
+
 ## Mandatory Rules
 
 1. **Never mock, stub, or fake anything.** Real implementations only.
 2. **TypeScript strict mode** — no `any`, no `@ts-ignore`.
 3. **Always kill terminals** after commands complete.
-4. **Always commit and push** as `nirholas`.
+4. **Always commit and push** as `preciousnwankwo`.
+5. **Never commit `.env` files, session cookies, or API keys.**
+6. **Never auto-post content. The user posts all content manually.**
\ No newline at end of file
diff --git a/config/hermes-mcp-config.json b/config/hermes-mcp-config.json
new file mode 100644
index 00000000..e31c3986
--- /dev/null
+++ b/config/hermes-mcp-config.json
@@ -0,0 +1,12 @@
+{
+  "mcpServers": {
+    "xactions": {
+      "command": "npx",
+      "args": ["-y", "xactions-mcp"],
+      "env": {
+        "XACTIONS_SESSION_COOKIE": "${XACTIONS_SESSION_COOKIE}",
+        "XACTIONS_MODE": "local"
+      }
+    }
+  }
+}
diff --git a/docs/hermes-integration.md b/docs/hermes-integration.md
new file mode 100644
index 00000000..1160d6f9
--- /dev/null
+++ b/docs/hermes-integration.md
@@ -0,0 +1,64 @@
+# Hermes Agent Integration
+
+XActions integrates with [Hermes Agent](https://hermes-agent.nousresearch.com/) for AI-powered Twitter content generation.
+
+## Overview
+
+1. **Hermes generates content** — security threads, vulnerability breakdowns, exploit analysis
+2. **Deslop pass** — all output is stripped of AI patterns (no em dashes, hedging, AI vocabulary)
+3. **User posts manually** — Hermes sends content to the user via Telegram. The user copies and posts it.
+
+No X API keys. No session cookies. No browser automation. Just content generation.
+
+## Quick Start
+
+### 1. Generate Content
+
+Hermes uses the security content pool at `~/.hermes/scripts/security-content-generator.py` to generate threads. Content is automatically deslopped.
+
+### 2. Receive Content
+
+Hermes sends content to the user via Telegram, 3 times daily:
+- Morning (9:00 UTC)
+- Afternoon (14:00 UTC)
+- Evening (22:37 UTC)
+
+### 3. Post Manually
+
+The user copies the content and posts it to X/Twitter manually. No automation.
+
+## Content Pipeline
+
+```
+Security Content Pool (26+ threads, auto-refreshed weekly)
+        ↓
+   Hermes picks topic (no repeats)
+        ↓
+   Deslop transformation
+        ↓
+   Hermes sends to user (Telegram)
+        ↓
+   User copies and posts manually
+        ↓
+   Log sent thread (no duplicates)
+```
+
+## Content Types
+
+- **Single tweet** — one vulnerability insight, under 280 chars
+- **Thread** — 3-8 tweets breaking down an exploit, with hook + value + CTA
+
+## Safety
+
+- **No auto-posting** — the user posts all content manually
+- **No duplicate content** — sent threads are logged and never repeated
+- **No secrets in repo** — `.env` is in `.gitignore`
+- **No X API keys needed** — no automation means no keys
+
+## Files
+
+| File | Purpose |
+|------|---------|
+| `skills/hermes-integration/SKILL.md` | Full integration guide |
+| `scripts/hermes-poster.js` | Format content for manual posting |
+| `.env.example` | Environment template (for future automation) |
diff --git a/scripts/hermes-poster.js b/scripts/hermes-poster.js
new file mode 100644
index 00000000..2f1843e4
--- /dev/null
+++ b/scripts/hermes-poster.js
@@ -0,0 +1,95 @@
+#!/usr/bin/env node
+// Hermes Content Poster — Format deslopped content for manual posting
+// Usage: node hermes-poster.js [content-file]
+// Reads a content file, splits it into tweet-sized chunks, outputs copy-paste ready text
+
+import fs from 'node:fs';
+import path from 'node:path';
+
+const CONTENT_DIR = process.env.HERMES_CONTENT_DIR || `${process.env.HOME}/.hermes/cron/output`;
+
+function parseContent(raw) {
+  const lines = raw.trim().split('\n');
+  const title = lines[0].replace(/[:\-–—].*$/, '').trim();
+  const body = lines.slice(1).join('\n').trim();
+  const tweets = splitIntoTweets(body);
+  return { title, body, tweets };
+}
+
+function splitIntoTweets(text) {
+  const maxLen = 280;
+  const tweets = [];
+  const sentences = text.match(/[^.!?]+[.!?]+/g) || [text];
+
+  let current = '';
+  for (const sentence of sentences) {
+    if ((current + ' ' + sentence).trim().length <= maxLen) {
+      current = (current + ' ' + sentence).trim();
+    } else {
+      if (current) tweets.push(current);
+      if (sentence.length <= maxLen) {
+        current = sentence.trim();
+      } else {
+        const words = sentence.split(' ');
+        current = '';
+        for (const word of words) {
+          if ((current + ' ' + word).length <= maxLen) {
+            current = (current + ' ' + word).trim();
+          } else {
+            if (current) tweets.push(current);
+            current = word;
+          }
+        }
+      }
+    }
+  }
+  if (current) tweets.push(current);
+
+  if (tweets.length > 1) {
+    return tweets.map((t, i) => `${i + 1}/${tweets.length} ${t}`);
+  }
+  return tweets;
+}
+
+function main() {
+  const inputFile = process.argv[2];
+  if (!inputFile) {
+    const files = fs.readdirSync(CONTENT_DIR)
+      .filter(f => f.startsWith('security-thread_') && f.endsWith('.txt'))
+      .sort().reverse().slice(0, 10);
+
+    if (files.length === 0) {
+      console.log('No content files found.');
+      process.exit(1);
+    }
+
+    console.log('Available content files:');
+    files.forEach((f, i) => console.log(`  ${i + 1}. ${f}`));
+    console.log('\nUsage: node hermes-poster.js <filename>');
+    process.exit(0);
+  }
+
+  const filePath = path.isAbsolute(inputFile) ? inputFile : path.join(CONTENT_DIR, inputFile);
+  if (!fs.existsSync(filePath)) {
+    console.error(`File not found: ${filePath}`);
+    process.exit(1);
+  }
+
+  const raw = fs.readFileSync(filePath, 'utf-8');
+  const { title, tweets } = parseContent(raw);
+
+  console.log(`\n${'='.repeat(60)}`);
+  console.log(`Title: ${title}`);
+  console.log(`Tweets: ${tweets.length}`);
+  console.log(`${'='.repeat(60)}\n`);
+
+  tweets.forEach((t, i) => {
+    console.log(`--- Tweet ${i + 1} (${t.length} chars) ---`);
+    console.log(t);
+    console.log();
+  });
+
+  console.log('Copy each tweet above and post to X/Twitter manually.');
+}
+
+main();
diff --git a/skills/hermes-integration/SKILL.md b/skills/hermes-integration/SKILL.md
new file mode 100644
index 00000000..98d8a14a
--- /dev/null
+++ b/skills/hermes-integration/SKILL.md
@@ -0,0 +1,71 @@
+---
+name: hermes-integration
+description: Integrates XActions with Hermes Agent for AI-powered Twitter/X content generation. Hermes generates deslopped security content and sends it to the user for manual posting. No automation of posting — the user posts themselves. Use when the user wants to generate Twitter content through Hermes or manage the content pipeline from Hermes to X/Twitter.
+license: MIT
+metadata:
+  author: fisher
+  version: "1.1"
+---
+
+# Hermes Integration
+
+Bridge between Hermes Agent and XActions for Twitter/X content generation.
+
+## How It Works
+
+1. **Hermes generates content** — security threads, vulnerability breakdowns, exploit analysis
+2. **Deslop pass** — all output is run through deslop (no em dashes, AI vocabulary, hedging)
+3. **User posts manually** — Hermes sends the content to the user. The user copies and posts it themselves.
+
+That's it. No X API keys, no session cookies, no browser automation. Just content generation.
+
+## Content Pipeline
+
+### Step 1: Generate
+
+Hermes picks a topic from the security content pool (`~/.hermes/scripts/security-content-generator.py`) or generates a fresh take on a recent exploit.
+
+Content types:
+- **Single tweet** — one vulnerability insight, under 280 chars
+- **Thread** — 3-8 tweets breaking down an exploit, with hook + value + CTA
+
+### Step 2: Deslop
+
+All content passes through the deslop pipeline before being shown to the user:
+- Zero em dashes
+- Contractions everywhere
+- Varied sentence rhythm
+- No AI vocabulary (robust, leverage, navigate, etc.)
+- Opinions allowed
+- Rough edges preserved
+
+### Step 3: User Posts
+
+Hermes sends the content to the user via Telegram. The user:
+- Copies the content
+- Pastes it into X/Twitter
+- Posts it manually
+
+No automation. No API keys needed. The user stays in full control.
+
+## Content Storage
+
+- Generated content: `~/.hermes/cron/output/security-thread_*.txt`
+- Sent threads log: `~/.hermes/scripts/sent-threads.json` (prevents duplicates)
+
+## Workflow Files
+
+| File | Purpose |
+|------|---------|
+| `skills/hermes-integration/SKILL.md` | This file — integration guide |
+| `skills/content-posting/SKILL.md` | Posting scripts and selectors (for reference) |
+| `skills/viral-thread-generation/SKILL.md` | Thread structure and hooks |
+| `skills/engagement-interaction/SKILL.md` | Like, reply, bookmark automation |
+| `skills/algorithm-cultivation/SKILL.md` | Feed training and growth |
+
+## Safety Rules
+
+1. **Never auto-post.** The user posts all content manually.
+2. **Never commit `.env` files or session cookies** (for when automation is added later)
+3. **Always deslop content before showing the user**
+4. **Log all sent content to prevent duplicates**