- Home
- Skills
- Writing & Content
- Blog Machine (Free Starter)
Works with the AI Agents you already use
Blog Machine (Free Starter)
Turns a keyword or rough idea into a post skeleton with section payoffs and a first sentence. Add a voice line to match your tone, or paste an outline to find its weak sections. One skill, free.
Free
Blog Machine (Free Starter)
Example session with this skill installed
Topic: onboarding emails for a SaaS free trial. Voice line: "I tested six trial flows last quarter. Four of them lost people on day two."
- Read your context and instructions
- Compiled the blog machine free
Example output (an illustration of what the skill returns, not a customer session).
SKELETON: What to Send on Day Two of a SaaS Trial
Reader asking: why do trial users go quiet on day two, and what should the email say?
Promise: by the end you will have one day-two email that gives users a reason to come back.
1. The day-two drop: pays off with where trial users actually go quiet (guess)
2. What the first email already did: pays off with a check of what day one promised
3. The one action to ask for: pays off with a single next step, not a feature tour
4. A day-two email you can adapt: pays off with a short template (guess)
First sentence: "Four of the six trial flows I tested lost people on day two, so day two is where this email earns its place." [voice: matched to your line: first person, short declarative sentences, a specific count up front]
Soft spots to confirm: the day-two claim, any numbers in section 1
What happened here
- The skill guessed first and tagged the two guessed sections, so you know where to push back.
- It stopped at four sections. The limit is five.
- The first sentence reused the shape of your voice line: first person, short, a count up front.
Next step
Edit the skeleton, then paste your edited outline back. The skill returns the same headings with weak sections tagged (no payoff, overlap, off-promise, too big, guess) and the three fixes to make first.
Want the full draft pipeline? Blog Machine (Persona, $59) adds outline-builder, draft-pipeline, seo-polisher and outline-template.
Connects securely to your tools. The creator never sees your data.
About this skill
Turns a keyword or rough idea into a post skeleton with section payoffs and a first sentence. Add two sentences you wrote and the first sentence matches your tone. Paste an outline and it marks the weak sections instead.
What is inside
One SKILL.md you can drop into any compatible agent. There are no other files in the package.
What it does
- Restates your topic as a reader and their question, then builds a promise sentence, four to five sections with a payoff each, and a working title.
- Tags every guess with (guess) so you know where to push back.
- Writes a first sentence, matched to a voice line if you give one.
- Checks a pasted outline and tags weak sections: no payoff, overlap, off-promise, too big, or guess.
Use cases
- Convert a brief client one-liner into a structured content plan.
- Quickly unstick a blog post that you are struggling to start.
- Prepare a rough content flow immediately before a client discovery call.
- Check your own outline before you start the draft.
Known limitations
Limited to five sections to maintain focus. It does not write the full post, check facts or research keywords. It makes assumptions that you need to confirm, and it marks each one as a guess.
Maximum of five sections per skeleton. Requires manual verification of AI "guesses" for technical or highly niche topics.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
This skill has been used 14 times.
Trust & safety
Security scan
Security scan passed today
- Free with an Agensi account
Permissions
No special permissions declared or detected