Content Pipeline Skill
This skill runs the full verifiable content generation pipeline. It takes a topic, generates a researched and polished article using local Ollama for drafting and EigenAI for cryptographic verification, then stores the proof on EigenDA.
When to use this skill
Use this skill when the user asks you to:
- •Write a technical article or blog post
- •Generate verified content about a topic
- •Run the content pipeline
- •Create a new article
How to run
- •
Extract the topic from the user's message. If no topic is specified, ask for one.
- •
Run the pipeline script:
node /Users/zeeshan/Downloads/eigen-openclaw/src/pipeline.js "<TOPIC>"
- •
The pipeline will run 4 agents in sequence:
- •Agent 1 (SEO Researcher) — runs on local Ollama (free), outputs keyword research and outline
- •Agent 2 (Draft Writer) — runs on local Ollama (free), outputs a 2000+ word article draft
- •Agent 3 (Verifier) — runs on EigenAI (uses free token credits), polishes and cryptographically verifies the article
- •Agent 4 (Proof Storage) — stores verification proof on EigenDA
- •
Once complete, the output is saved to a timestamped folder under:
code/Users/zeeshan/Downloads/eigen-openclaw/output/<timestamp>_<topic-slug>/
With these files:
- •
1-research.md— SEO research and outline - •
2-draft.md— Raw draft from Ollama - •
3-final-article.md— Final verified article - •
4-proof.json— Cryptographic proof metadata (EigenAI model, token usage, EigenDA commitment)
- •
- •
After the pipeline completes, read the
3-final-article.mdfile and share a summary with the user. Include:- •Word count
- •Whether it was verified by EigenAI
- •The EigenDA commitment (if available)
- •The output directory path
- •
Ask the user if they want to:
- •Publish it (use the
publish-articleskill) - •Edit it (make changes to the final article)
- •Regenerate it with a different angle
- •Publish it (use the
Prerequisites
Before running, ensure all services are healthy:
node /Users/zeeshan/Downloads/eigen-openclaw/src/health-check.js
Required services:
- •Ollama must be running locally (
ollama serve) - •EigenAI grant must be active (check at https://terminal.eigencloud.xyz)
- •EigenDA Proxy should be running at localhost:3100
Token Budget
Each article uses approximately 4,000-5,000 EigenAI tokens. The user has a 1M free token budget. Monitor remaining tokens in the pipeline output.