You follow 20 technical YouTube channels and spend two hours every week scanning for video ideas, researching angles, and writing script outlines — before any actual filming begins. None of that work requires your creative judgment. All of it can run automatically.
This tutorial is one of the most practical OpenClaw use cases for solo creators: a single agent that scouts trending ideas from a configurable topic list, selects the strongest candidate, writes a structured script outline, and logs the result to a content tracker. One run, one deliverable.
Test the research and outlining step before you write a single line of config. Paste a topic into wisgate.ai/studio/image with
claude-opus-4-6selected and see what the agent produces. Get your API key at wisgate.ai/hall/tokens when you're ready to schedule the first run.
AI YouTube Content Pipeline: What the Agent Does
The pipeline runs as a single OpenClaw agent conversation — manually or on a weekly cron schedule. Four steps per run:
| Step | Action | Output |
|---|---|---|
| Scout | Scans topic list for trending angles and underserved questions | 3–5 ranked candidate ideas |
| Select | Picks the strongest idea by search intent, freshness, channel fit | One title with rationale |
| Outline | Writes a structured script with hook, sections, timestamps, CTA | Markdown outline file |
| Track | Appends title, source, outline path, and date to tracker log | Updated content-tracker.md |
Components needed: OpenClaw configured with WisGate, the agent YAML in Section 3, and a topics.yaml file with your channel's focus areas.
OpenClaw API YouTube Content Automation: WisGate Setup
Step 1 — Open the configuration file
OpenClaw stores its configuration in a JSON file in your home directory. Open your terminal and edit:
nano ~/.openclaw/openclaw.json
Step 2 — Add the WisGate provider to your models section
Copy and paste the following into your models section. This defines WisGate as a custom provider pointing to the Claude Opus model:
"models": {
"mode": "merge",
"providers": {
"moonshot": {
"baseUrl": "https://api.wisgate.ai/v1",
"apiKey": "WISGATE-API-KEY",
"api": "openai-completions",
"models": [
{
"id": "claude-opus-4-6",
"name": "Claude Opus 4.6",
"reasoning": false,
"input": ["text"],
"cost": {
"input": 0,
"output": 0,
"cacheRead": 0,
"cacheWrite": 0
},
"contextWindow": 256000,
"maxTokens": 8192
}
]
}
}
}
Replace WISGATE-API-KEY with your key from wisgate.ai/hall/tokens. Confirm claude-opus-4-6 pricing at wisgate.ai/models.
Step 3 — Save and restart
Ctrl + O→Enterto saveCtrl + Xto exitCtrl + Cto stop the current session, then runopenclaw tui
Note: OpenClaw was previously known as ClawdBot and MoltBot — these steps apply to all versions.
LLM Video Content Research Automation: The Agent YAML
Copy this directly into your OpenClaw agents directory:
# youtube-content-pipeline.yaml
name: YouTube Content Pipeline
description: Scout ideas, select strongest, outline script, log to tracker
schedule: "" # Set "0 9 * * 1" for weekly Monday 09:00
model:
provider: custom
base_url: "https://api.wisgate.ai/v1"
model_id: "claude-opus-4-6"
api_key: "${WISDOM_GATE_KEY}"
config:
topics_file: "./topics.yaml"
output_dir: "./outlines"
tracker_file: "./content-tracker.md"
system_prompt: |
You are a YouTube content pipeline agent for a technical developer channel.
STEP 1 — SCOUT
Read the topic list. For each topic, identify 1–2 current angles that are
underserved in search, relevant to developers, and fresh (last 30–60 days).
Return 3–5 ranked candidate ideas with one-sentence rationale each.
STEP 2 — SELECT
Choose the single strongest candidate. State:
- Working title (optimized for search intent)
- Why this idea: search demand, freshness, channel fit
- Estimated search signal: High / Medium / Low
STEP 3 — OUTLINE
Produce a Markdown script outline:
- Hook (0:00–0:45): one specific question or claim that creates urgency
- Section 1 (0:45–3:00): context and problem setup
- Section 2 (3:00–7:00): core technical explanation or demo
- Section 3 (7:00–10:30): implementation or comparison
- CTA (10:30–11:00): subscribe ask + one follow-up video recommendation
Each section: 3–5 bullet points. Technical specificity only — no filler.
STEP 4 — TRACK
Append one line to content-tracker.md:
| [DATE] | [TITLE] | [TOPIC SOURCE] | [OUTLINE PATH] | Outlined |
max_tokens: 4096
Populate topics.yaml with your channel's focus areas:
topics:
- name: "LLM API integration"
subtopics: ["multi-model routing", "token cost optimization", "streaming"]
- name: "Developer tooling"
subtopics: ["AI-assisted debugging", "code review agents", "local LLM setup"]
- name: "Self-hosted infrastructure"
subtopics: ["home server automation", "Docker", "monitoring stacks"]
OpenClaw Use Cases: Your Weekly Content Calendar, Automated
After four weeks, your content-tracker.md looks like this:
| Date | Working Title | Source | Outline | Status |
|------------|------------------------------------------------|-------------------|--------------------------|----------|
| 2026-03-10 | How to Route LLM Calls by Cost Without a Framework | LLM API integration | ./outlines/2026-03-10.md | Outlined |
| 2026-03-17 | Self-Hosted Monitoring Stack Under $5/Month | Self-hosted infra | ./outlines/2026-03-17.md | Recording|
| 2026-03-24 | Claude vs GPT for Code Review: 30-Day Test | Developer tooling | ./outlines/2026-03-24.md | Editing |
| 2026-03-31 | Token Cost Optimization: 6 Patterns That Work | LLM API integration | ./outlines/2026-03-31.md | Outlined |
The tracker becomes your content calendar. Update the Status column as videos move from Outlined → Recording → Editing → Published. The pipeline handles everything before that first column — every week, automatically.
The YAML is ready to copy. Your channel list is the only variable. Get your WisGate API key at wisgate.ai/hall/tokens — trial credits included. Before setting the schedule, paste the system prompt and one topic into wisgate.ai/studio/image and check the outline quality against your standards. That's the only step left.