diff --git a/README.md b/README.md
index 18ab938..e770aff 100644
--- a/README.md
+++ b/README.md
@@ -64,6 +64,7 @@ Skills work seamlessly with **Command Code**  providing consistent, high-quality
 
 ### Content & Communication
 
+- **[Memory Engine](skills/memory-engine/)** - Store and retrieve long-term knowledge across sessions with semantic naming and hybrid search. Use when you need to remember decisions, recall past context, or search what was discussed or built in a previous session.
 - **[Content Research Writer](skills/content-research-writer/)** - Assists in writing high-quality content by conducting research, adding citations, improving hooks, and providing section-by-section feedback.
 - **[Internal Comms](skills/internal-comms/)** - Helps write internal communications including 3P updates, company newsletters, FAQs, status reports, and project updates using company-specific formats.
 - **[Meeting Insights Analyzer](skills/meeting-insights-analyzer/)** - Analyzes meeting transcripts to uncover behavioral patterns including conflict avoidance, speaking ratios, filler words, and leadership style.
@@ -98,6 +99,7 @@ Skills work seamlessly with **Command Code**  providing consistent, high-quality
 
 ### Workflow Automation
 
+- **[Site Automation](skills/site-automation/)** - Manage website content via automation APIs. Sync GitHub projects, generate changelogs, create posts, send newsletters, and manage comments. Use for any website content management task.
 - **[File Organizer](skills/file-organizer/)** - Intelligently organizes files and folders by understanding context, finding duplicates, and suggesting better organizational structures.
 - **[Invoice Organizer](skills/invoice-organizer/)** - Automatically organizes invoices and receipts for tax preparation by reading files, extracting information, and renaming consistently.
 - **[Raffle Winner Picker](skills/raffle-winner-picker/)** - Randomly selects winners from lists, spreadsheets, or Google Sheets for giveaways and contests with cryptographically secure randomness.
diff --git a/skills/memory-engine/SKILL.md b/skills/memory-engine/SKILL.md
new file mode 100644
index 0000000..0053f15
--- /dev/null
+++ b/skills/memory-engine/SKILL.md
@@ -0,0 +1,129 @@
+---
+name: memory-engine
+description: >
+  Store and retrieve long-term knowledge across sessions with semantic naming and hybrid search.
+  Use when you need to remember something for the future, recall past decisions, look up prior context,
+  or search what was discussed or built in a previous session.
+user-invocable: true
+---
+
+# Memory Engine 🧠
+
+Persistent, searchable memory across all sessions. Storage uses semantic file naming and hybrid search (semantic + keyword fallback).
+
+## When to Use This Skill
+
+- "Remember that…" / "Don't forget…" → store a memory
+- "What did we decide about…" / "What have we talked about…" → search memory
+- "What's the status of…" → search memory
+- Start of any new session — search relevant topics before answering
+- After completing important work — store a summary immediately
+
+## How It Works
+
+**Storage:** Memory files use semantic naming convention:
+```
+YYYY-MM-DD__[CATEGORY]__[TOPIC]__[CONTEXT].md
+```
+
+**Search:** Hybrid approach:
+1. Try semantic search first (embeddings-based)
+2. Expand query with synonyms if sparse results
+3. Fall back to keyword search if semantic unavailable
+4. Return diagnostics showing which methods were used
+
+## Commands
+
+### Store a Memory
+```bash
+memory_store "[fact or summary]" \
+  --category [briefing|strategy|audit|vision|code|project|bug|feature|research|decision|general] \
+  --topic [your-topic] \
+  --context [optional-descriptor]
+```
+
+### Search Memory (Enhanced)
+```bash
+memory_search "[query]" --limit 10 --diagnostics
+```
+
+### Search Memory (Basic Fallback)
+```bash
+memory_search_basic "[query]" --limit 10
+```
+
+### Check Memory Health
+```bash
+memory_health
+```
+
+## File Naming Convention
+
+Memory files follow a semantic naming pattern:
+
+```
+YYYY-MM-DD__[CATEGORY]__[TOPIC]__[CONTEXT].md
+```
+
+**Categories:** briefing, strategy, audit, vision, code, project, bug, feature, research, decision, general
+
+**Examples:**
+- `2026-02-25__briefing__morning-signal__daily.md`
+- `2026-02-24__strategy__architecture__design.md`
+- `2026-02-24__audit__database__schema.md`
+- `2026-02-20__decision__authentication__oauth.md`
+
+## Examples
+
+```bash
+# Store a decision
+memory_store \
+  "Decided to use semantic file naming with YYYY-MM-DD__CATEGORY__TOPIC__CONTEXT pattern" \
+  --category decision --topic memory-engine --context naming
+
+# Store a briefing summary
+memory_store \
+  "Morning briefing sent successfully to 142 subscribers with 8 key insights" \
+  --category briefing --topic daily-briefing --context summary
+
+# Search for past decisions
+memory_search "authentication oauth" --limit 5
+
+# Exploratory search with diagnostics
+memory_search "what have we decided about architecture" --limit 5 --diagnostics
+```
+
+## Search Behavior
+
+The enhanced search engine automatically:
+
+1. **Detects query type** (factual / conceptual / exploratory) and adjusts relevance threshold
+2. **Tries semantic search** first using embeddings
+3. **Expands query** with synonyms if results are sparse
+4. **Falls back to keyword search** if semantic search is unavailable or returns nothing
+5. **Returns diagnostics** showing which methods were used and why
+
+## Guardrails
+
+- **Always store** important facts, decisions, and task completions immediately after they happen
+- **Always search** before answering questions about past sessions — never assume context is absent
+- Use `--diagnostics` when a search returns no results to understand why
+- If enhanced search is unavailable, fall back to basic keyword search
+- Store the **summary of significant changes**, not just raw data — make it human-readable for future retrieval
+- Keep memory entries concise but complete (50-500 words)
+- Use consistent category and topic names across sessions for better search results
+
+## Integration
+
+This skill works best when:
+- Invoked at the start of each session to recall relevant context
+- Used after completing major work to store summaries
+- Combined with other skills that need historical context
+- Integrated into agent workflows for persistent learning
+
+## Technical Notes
+
+- Semantic search requires embeddings API (e.g., Google Generative AI)
+- Keyword search works offline without external dependencies
+- Memory files are stored as markdown for human readability
+- Search results are scored by relevance and recency
diff --git a/skills/site-automation/SKILL.md b/skills/site-automation/SKILL.md
new file mode 100644
index 0000000..50ea178
--- /dev/null
+++ b/skills/site-automation/SKILL.md
@@ -0,0 +1,198 @@
+---
+name: site-automation
+description: >
+  Manage website content via automation APIs. Sync GitHub projects, generate changelogs,
+  create posts, send newsletters, manage comments, and check health. Use when any task
+  involves website content management or newsletter operations.
+user-invocable: true
+---
+
+# Site Automation 🌐
+
+Manage website content through automation APIs. Sync GitHub projects, generate changelogs, create posts, send newsletters, and manage comments.
+
+## When to Use This Skill
+
+- "Sync the latest GitHub projects"
+- "Generate a changelog from recent commits"
+- "Create a blog post"
+- "Send a newsletter"
+- "Check recent comments on posts"
+- "Reply to a comment"
+- "List newsletter subscribers"
+
+## Prerequisites
+
+Set up environment variables:
+```bash
+export SITE_BASE_URL="http://localhost:3000"  # or your site URL
+export CLAW_AUTOMATION_KEY="your-api-key"     # Bearer token for API auth
+export GITHUB_USERNAME="your-github-user"     # GitHub user for project sync
+```
+
+Store these in your `.env` file for persistence.
+
+## Commands
+
+### Health Check
+```bash
+site_automation health
+```
+
+### Full Sync (Projects + Changelog)
+```bash
+site_automation full-sync
+```
+
+Syncs GitHub projects to database and generates changelog from recent commits.
+
+### Sync Projects Only
+```bash
+site_automation sync-projects
+```
+
+Pulls latest GitHub repositories and updates project database.
+
+### Sync Changelog Only
+```bash
+site_automation sync-changelog
+```
+
+Generates changelog entries from recent git commits.
+
+### Create a Blog Post
+```bash
+site_automation create-post \
+  --slug post-slug \
+  --title "Post Title" \
+  --content-file /path/to/content.md \
+  --type blog \
+  --tags tag1 tag2 tag3
+```
+
+### Create a Changelog Post
+```bash
+site_automation create-post \
+  --slug changelog-slug \
+  --title "Changelog Title" \
+  --content-file /path/to/content.md \
+  --type changelog
+```
+
+### Send Newsletter
+```bash
+site_automation send-newsletter \
+  --subject "Newsletter Subject" \
+  --html-file /path/to/email.html \
+  --idempotency-key unique-key-for-deduplication
+```
+
+### List Subscribers
+```bash
+site_automation list-subscribers
+```
+
+### List Recent Comments
+```bash
+site_automation list-comments --since 2026-02-20
+```
+
+### Reply to a Comment
+```bash
+site_automation reply-comment \
+  --post-id POST_UUID \
+  --body "Your reply text" \
+  --source-comment-id COMMENT_UUID
+```
+
+## API Endpoints (Direct)
+
+If you need to call the API directly:
+
+| Method | Endpoint | Purpose |
+|--------|----------|---------|
+| GET | `/api/automation/health` | Health check |
+| POST | `/api/automation/projects/sync` | GitHub → projects DB |
+| POST | `/api/automation/changelog/github` | Commits → changelog posts |
+| POST | `/api/automation/posts` | Create blog/changelog post |
+| GET | `/api/automation/comments` | Poll recent comments |
+| POST | `/api/automation/comments/respond` | Reply to comment |
+| GET | `/api/automation/newsletter/subscribers` | List subscribers |
+| POST | `/api/automation/newsletter/send` | Send newsletter |
+
+**Auth header:** `Authorization: Bearer <CLAW_AUTOMATION_KEY>`
+
+## Workflow: Publish Article from Intelligence
+
+1. Generate article content (from briefing, research, or manual writing)
+2. Write content to a temporary `.md` file
+3. Call `create-post` with the file
+4. Optionally send newsletter linking to the new post
+
+## Examples
+
+```bash
+# Full sync before publishing
+site_automation full-sync
+
+# Create a blog post
+site_automation create-post \
+  --slug "ai-trends-2026" \
+  --title "AI Trends in 2026" \
+  --content-file ./article.md \
+  --type blog \
+  --tags ai trends 2026
+
+# Send newsletter
+site_automation send-newsletter \
+  --subject "Weekly Digest — Feb 25, 2026" \
+  --html-file ./newsletter.html \
+  --idempotency-key "digest-2026-02-25"
+
+# Check recent comments
+site_automation list-comments --since 2026-02-20
+
+# Reply to a comment
+site_automation reply-comment \
+  --post-id "550e8400-e29b-41d4-a716-446655440000" \
+  --body "Thanks for the feedback!" \
+  --source-comment-id "6ba7b810-9dad-11d1-80b4-00c04fd430c8"
+```
+
+## Integration Patterns
+
+### With Briefing Pipeline
+The daily briefing automatically runs `full-sync` before synthesizing the newsletter. This ensures:
+- New GitHub projects are synced to the database
+- Recent commits become changelog entries
+- Subscriber list is current before send
+
+### With Content Pipeline
+Use in content creation workflows:
+1. Research/write content
+2. Store in temporary file
+3. Call `create-post` to publish
+4. Trigger newsletter send if needed
+
+### With Comment Moderation
+Monitor comments and respond:
+1. Poll recent comments with `list-comments`
+2. Review for moderation
+3. Reply with `reply-comment` or flag for manual review
+
+## Response Format
+
+After execution, report:
+- Counts (synced, created, sent)
+- Any errors or warnings
+- Next steps if applicable
+
+## Guardrails
+
+- Always run `health` check before major operations
+- Use idempotency keys for newsletter sends to prevent duplicates
+- Validate markdown content before posting
+- Store API key securely (never hardcode)
+- Test with `--draft` flag before publishing
+- Monitor rate limits on GitHub API calls
+- Keep changelog entries concise and user-focused