Blog Post AI workflow template: A blog post, brief to publish-ready.

The Blog Post template makes your AI take a post from brief to publish-ready draft: pin the topic, reader, angle and call to action, build an outline, write the draft, edit it against the brief, then do a final read-through. You approve the outline before drafting and the post before publishing.

6 steps · 2 approvals · 1 loop back

Blog Post · example runyour AI client
  1. 01Brief
  2. 02Outlineyou approve
  3. 03Draft
  4. 04Edit
  5. 05Final Reviewyou approve
  6. 06Completeclosing note
task created · workflow attached

example · the steps are this template's real steps

in short

What does the Blog Post template do?

For marketers and founders who write with an AI and want the outline agreed before a full draft exists.

It is a stored workflow, not a prompt. Your AI client receives one step at a time from ConvOps over MCP, does the work, and advances. The engine holds the run at each of its two approvals until a person says yes.

  • Write a product-announcement post from a one-line brief to a publish-ready draft
  • Turn a rough idea into a structured article with a working title and clear sections
  • Edit a finished draft against the brief before it goes live
step by step

How does the Blog Post workflow run?

6 steps, in this order. Below each one is the instruction your AI receives at that step, exactly as the template stores it.

Brief
Outlineapproval
Draft
Edit
Final Reviewapproval
Completeclosing note
  1. 01

    Brief

    the agent works

    Establish the brief for this post. Do not start writing yet.

    Capture, in plain prose:

    • Topic: what the post is about, in one sentence
    • Target reader: who this is for and what they already know
    • Angle: the specific take or insight, not a generic overview
    • Goal: what the reader should think, feel, or do after reading
    • Primary call to action: the one next step you want

    If any of these is missing or vague, ask the operator before advancing. A weak brief produces a weak post.

    Before advancing, save the full brief as a task note with task_notes_add. The next step (Outline) reads this note as its named input.

  2. 02

    Outline

    waits for your approval

    Read the task notes with task_notes_list: the note from the Brief step contains topic, reader, angle, goal, and CTA. Work from it.

    Build the outline. What good looks like:

    • Working title: specific, not a placeholder
    • Hook: the opening line or idea that earns the next sentence
    • H2 section headings in order, each with a one-line summary of what that section argues or delivers
    • Target reader restated so the structure stays aimed at them
    • Closing CTA matching the brief

    Keep the outline tight: a reader should grasp the whole arc from the headings alone.

    Save the full outline as a task note with task_notes_add. Then ask the operator to approve the outline before any drafting begins. This is a checkpoint: the AI cannot advance until you approve.

    no rejection path · the run moves on only when approved

  3. 03

    Draft

    the agent works

    Read the task notes with task_notes_list: the note from the Outline step contains the approved structure. Work from it.

    Write the full first draft following the approved outline section by section. Hold to the angle and the target reader from the brief. Use concrete examples over abstractions. Write the real hook, not a placeholder. Do not pad to hit a length: every paragraph earns its place.

    Before advancing, save the full draft as a task note with task_notes_add so the Edit step (or a fresh session) picks up exactly here. If the draft exceeds 10,000 characters, split it across sequential notes or save it where the operator keeps documents and record that location in the note.

  4. 04

    Edit

    the agent works

    Read the task notes with task_notes_list: the note from the Draft step contains the full first draft, and the Brief note holds the goal and CTA to check against. Work from them.

    Edit the draft for:

    • Clarity: cut filler, fix any sentence that needs a second read
    • Flow: each section leads into the next; the hook still lands
    • Accuracy: every claim is correct and specific
    • Brief alignment: the angle, target reader, and CTA all hold
    • Title: confirm it is the strongest accurate option

    Save the full edited draft as a task note with task_notes_add, replacing the working draft as the current version. The Final Review step reads this note.

  5. 05

    Final Review

    waits for your approval

    Read the task notes with task_notes_list: the note from the Edit step contains the current version. Work from it.

    Do a final read-through as the target reader would. Confirm the title, hook, sections, and CTA all deliver on the brief and that nothing is broken or unfinished. Surface anything still off.

    Ask the operator to approve the post for publishing. This is a checkpoint: publishing is the irreversible moment, so the AI cannot advance until you approve. If you reject, the workflow returns to Edit.

    rejected → back to Edit

  6. 06

    Complete

    closes with a note

    The post is approved and ready to publish. Confirm the final approved draft is saved as a task note so the published version stays on the record, then close out the task.

where a person decides

Where do you approve in Blog Post?

Two steps wait for a person. You answer in the chat; the engine records the verdict and the audit row records who gave it.

step 2

Outline

Fixing structure in an outline is cheap; fixing it in a full draft is not. You approve the title, hook and sections before drafting begins.

what you review

  • Working title: specific, not a placeholder
  • Hook: the opening line or idea that earns the next sentence
  • H2 section headings in order, each with a one-line summary of what that section argues or delivers
  • Target reader restated so the structure stays aimed at them
  • Closing CTA matching the brief

no rejection path · approve to continue

step 5

Final Review

Publishing is the irreversible moment. The AI reads the post as the target reader would and surfaces anything off. Reject it and the run returns to Edit.

reject → back to Edit

gates the engine checks without you

  • Complete closes only when your AI leaves a closing note. The note is saved to the task as context, so later work can find why it ended the way it did.
what you need

What do you need to run Blog Post?

Read from the template's own steps: the tools they call, the files they write, and whether they touch code.

An AI client
Claude Code, Claude (desktop, web, Cowork), OpenAI Codex, ChatGPT, Cursor, OpenCode, or any MCP client. ConvOps does not run the model; your client does the work.
ConvOps
A workspace with Blog Post installed. It attaches to the task task type, so new work of that type starts on it.
Tools your AI calls
workflow_advance to move between steps, plus task_notes_add and task_notes_list to save and read each step's output.
Where the work lands
Task notes in ConvOps. Each step saves its output with task_notes_add, and the next step reads it back.
A repository
No. A plain chat client is enough.
Roles named in steps
None. One AI runs every step.
make it yours

How do you customize Blog Post?

An installed template is a workflow in your workspace. Ask your AI to change it in plain words; the edit lands on the one step that needs it.

ChatGPTexample

new chat

more example prompts

  • "Add an SEO step after Edit that checks the title, headings and meta description."
  • "In Brief, always load our brand voice guide."
  • "Add an approval to Draft so I read the first draft before the edit."
the difference

Blog Post: without a workflow vs with it

  • A post gets drafted with no outline, wanders off topic, and never matches the brief you started from
  • You close the chat halfway through the draft and lose the whole thread of where the piece was going
  • Nobody can tell which draft is current or what changed between the outline and the published version
  • The process lives in a prompt, and each run follows it a little differently
questions

Blog Post template FAQ

Can I close the chat mid-post and come back later?

Yes. Every step saves its output as a task note (the brief, outline, draft, edit), and each next step reads the previous note, so a fresh session picks up exactly where the last one stopped.

What does the Edit step check?

Clarity, flow, accuracy of every claim, alignment with the brief (angle, reader, call to action), and whether the title is the strongest accurate option.

Does it publish the post?

No. The workflow ends with an approved, publish-ready draft saved as a task note. You or your AI client publish it to your CMS.

Which AI clients can run the Blog Post template?

Any MCP client connected to ConvOps. Verified clients: Claude Code, Claude (desktop, web, Cowork), OpenAI Codex, ChatGPT, Cursor, OpenCode. It needs no repository, so a plain chat client works.