diff --git a/blog/ideas/001-why-backing-up-an-agent-is-so-hard.md b/blog/ideas/001-why-backing-up-an-agent-is-so-hard.md
deleted file mode 100644
index 441414a..0000000
--- a/blog/ideas/001-why-backing-up-an-agent-is-so-hard.md
+++ /dev/null
@@ -1,47 +0,0 @@
-# 001 — Why backing up an agent is so hard
-
-**Status:** Research in progress
-**Tags:** Technical, Backups, DevOps
-
-## Research notes
-
-### Real filesystem changes after one week of work on a Linux agent
-
-```
-504.6M ./usr
- 43.3M ./var
-448.9M ./root
- 54.0K ./etc
-996.8M .
-```
-
-Most users only back up the home directory (`/root` = 448.9M). But over half the changes live outside it — `/usr` (504.6M) and `/var` (43.3M) contain installed packages, pip/npm global installs, database files, logs, and system configs. A naive home-dir-only backup misses 55% of what changed.
-
-At restore time, you get your config back but the agent won't run — missing dependencies, missing system packages, missing database files.
-
-### What this means for the post
-
-- Agent environments are not just config files — they're full Linux systems with packages, services, and state scattered everywhere
-- "Backup your home directory" is dangerously incomplete advice for AI agents
-- Derez.ai's full-disk snapshot approach is the right solution
-- This is a strong selling point: the agent works after restore, not just the config
-
-### Next data points
-
-Oliver will do two more memory analysis snapshots after running the speed-run setups (Odoo + Cold Email). Those will show how much additional state those integrations add.
-
-### Outline
-
-1. The sprawl problem — why agents are harder to back up than a standard server
- *Include the filesystem analysis table*
-2. What a real agent backup needs to capture
-3. How Borg/deduplicated snapshots solve it
-4. How Derez.ai does it automatically (one-click restore)
-5. Best practices for users
-6. Better save than sorry — the selling point
-
-### References
-
-- Borg backup docs
-- Hermes Agent directory structure (~/.hermes/)
-- Filesystem analysis: `du -sch /usr /var /root /etc` after one week
\ No newline at end of file
diff --git a/blog/ideas/002-cold-email-at-scale.md b/blog/ideas/002-cold-email-at-scale.md
deleted file mode 100644
index be30e42..0000000
--- a/blog/ideas/002-cold-email-at-scale.md
+++ /dev/null
@@ -1,24 +0,0 @@
-# 002 — Cold email at scale: from one-off to campaign
-
-**Status:** Idea
-**Tags:** Marketing, Cold Email, Automation
-
-## Research notes
-
-- We already have a cold email speed-run post — this would be the "next level"
-- Moving from single emails to sequenced campaigns
-- Follow-ups, A/B testing, reply detection
-- CRM integration for tracking
-
-## Outline
-
-1. Why one-off cold emails leave money on the table
-2. Building a sequence: intro → follow-up 1 → follow-up 2 → breakup
-3. Using Hermes Agent tools for reply detection and routing
-4. Measuring open rates, reply rates, conversion
-5. Tying into a CRM (Odoo?) for pipeline tracking
-
-## References
-
-- Existing cold email speed-run post
-- Alex Hormozi / Cold email playbooks
\ No newline at end of file
diff --git a/blog/ideas/003-odoo-beyond-speed-run.md b/blog/ideas/003-odoo-beyond-speed-run.md
deleted file mode 100644
index 1bb289e..0000000
--- a/blog/ideas/003-odoo-beyond-speed-run.md
+++ /dev/null
@@ -1,27 +0,0 @@
-# 003 — Odoo + AI agent: beyond the speed run
-
-**Status:** Idea
-**Tags:** Odoo, Integration, CRM
-
-## Research notes
-
-- The speed-run post shows a basic Odoo connection
-- This post goes deeper: actual workflows that save time
-- Inventory alerts, automated invoice follow-ups, lead scoring
-
-## Outline
-
-1. Recap: what the speed run covered
-2. Real use cases:
- - auto-follow-up on stale leads
- - inventory low-stock alerts via email/Slack
- - monthly sales report generated and emailed
-3. Combining Odoo tools with Hermes Agent skills
-4. Security: creating a restricted Odoo user for the agent
-5. Next steps for readers
-
-## References
-
-- Existing Odoo speed-run post
-- Odoo External API docs
-- Odoo4projects.com free trial
\ No newline at end of file
diff --git a/operations.html b/operations.html
index 75e0a33..c63a625 100644
--- a/operations.html
+++ b/operations.html
@@ -506,61 +506,6 @@
.goals-grid { grid-template-columns: 1fr; }
}
- /* ── Blog ideas ──────────────────────────── */
- .blog-idea-item {
- display: flex;
- gap: 16px;
- padding: 20px 24px;
- background: #0e0e14;
- border: 1px solid #1a1a2e;
- border-radius: 18px;
- transition: border-color 0.2s, transform 0.2s;
- align-items: flex-start;
- margin-bottom: 16px;
- }
- .blog-idea-item:hover {
- border-color: #6c8cff;
- transform: translateY(-2px);
- }
- .blog-idea-checkbox {
- flex-shrink: 0;
- padding-top: 3px;
- }
- .blog-idea-checkbox input[type="checkbox"] {
- accent-color: #6c8cff;
- width: 17px;
- height: 17px;
- cursor: default;
- }
- .blog-idea-body { flex: 1; min-width: 0; }
- .blog-idea-title {
- font-weight: 700;
- font-size: 16px;
- color: #e8e8f0;
- margin-bottom: 6px;
- }
- .blog-idea-desc {
- font-size: 13px;
- color: #888;
- line-height: 1.7;
- margin-bottom: 10px;
- }
- .blog-idea-meta {
- display: flex;
- gap: 6px;
- flex-wrap: wrap;
- }
- .blog-idea-tag {
- display: inline-block;
- padding: 3px 12px;
- font-size: 11px;
- font-weight: 600;
- border-radius: 12px;
- background: rgba(108,140,255,0.1);
- color: #6c8cff;
- border: 1px solid rgba(108,140,255,0.15);
- letter-spacing: 0.02em;
- }
@@ -703,62 +648,6 @@
-
-
-
-
📝 Blog Post Ideas
- tick when live
- ▼
-
-
-
-
-
-
-
-
#001 — Why backing up an agent is so hard
-
Modern AI agents install packages everywhere — pip in venvs, npm globally, configs scattered across /etc, ~/.config, /opt, custom paths. A naive filesystem backup misses half the environment.
-
- Technical
- Backups
- DevOps
-
-
-
-
-
-
-
-
-
#002 — Cold email at scale: from one-off to campaign
-
Moving from single cold emails to sequenced campaigns with follow-ups, A/B testing, reply detection, and CRM integration.
-
- Marketing
- Cold Email
- Automation
-
-
-
-
-
-
-
-
-
#003 — Odoo + AI agent: beyond the speed run
-
Deeper Odoo workflows: auto-follow-up on stale leads, low-stock alerts, monthly sales reports — all via the agent.