FerroxLabs Voice-Locked Blog Publishing
Estimated time: 30-45 minutes
This workflow takes one sharp angle and turns it into a published long-form blog post plus three platform derivatives, all locked to the author's voice. It moves from outline to draft to a humanize pass that strips AI tells, then fans the finished anchor out to email, LinkedIn, and an X-thread. Review gates sit at the outline, the draft, and the final bundle so the user steers before anything ships.
By the end you will have: a confirmed angle and reader profile, a structural outline, a humanized long-form draft, and three voice-aligned derivatives ready to publish.
When to Use
- User wants to publish a long-form blog post with derivatives
- User wants a repeatable outline-to-repurpose publishing process
- User wants platform derivatives that match the blog's voice
Workflow Steps
Step 1: Capture the Angle (uses: content-blog)
Ask the user for the angle in one sentence: the take, contrarian belief, or lived insight worth publishing. Do not proceed until you have a single, sharp angle. This is the spine every later step hangs on.
- Input: user's one-sentence angle
- Output: confirmed publishing angle
- Key focus: one sharp, specific take, not a broad topic
Step 2: Infer the Reader Profile (uses: sales-icp)
From the angle, infer the ideal reader profile (who this is for and what they already believe). State the inferred profile and your reasoning, then let the user correct it. Use the confirmed profile to calibrate tone and proof throughout the rest of the workflow.
- Input: confirmed angle from Step 1
- Output: reader/audience profile with reasoning
- Key focus: a concrete reader, validated by the user
Step 3: Build the Outline (uses: content-blog)
Produce an outline-only pass: the structural beats and the proof points that support the angle for this audience. Show the outline. Then ask whether to refine it or proceed to the full draft. If the user wants changes, revise the outline against their feedback and show it again, up to three iterations, before moving on.
- Input: angle, audience profile
- Output: approved blog outline with beats and proof points
- Key focus: structure and proof before prose
Step 4: Write the Full Draft (uses: content-blog)
Expand the approved outline into a complete long-form draft in the author's voice, holding the angle and the audience profile throughout.
- Input: approved outline, angle, audience profile
- Output: full long-form blog draft
- Key focus: a complete draft that delivers on the outline
Step 5: Humanize the Draft (uses: content-humanize)
Run the draft through a humanize pass: apply the voice profile from ~/wayland/voice-notes.md and strip AI tells. Show the humanized draft, then ask whether to refine or proceed to the repurpose chain. If the user wants changes, re-humanize against their feedback and show it again, up to three iterations.
- Input: full draft, voice profile at ~/wayland/voice-notes.md
- Output: voice-locked, humanized anchor draft
- Key focus: sounds like the author, no AI tells
Step 6: Repurpose to Derivatives (uses: content-repurpose)
From the humanized anchor, generate three derivatives for email, LinkedIn, and an X-thread, each voice-aligned and shaped for its platform. Show all three. Then ask whether to ship the bundle or refine a specific derivative; if a derivative needs work, revise just that one against the user's feedback, up to three iterations.
- Input: humanized anchor, angle, audience profile
- Output: email, LinkedIn post, and X-thread derivatives
- Key focus: platform-native shapes, one consistent voice
Step 7: Assemble the Publishing Bundle (uses: content-repurpose)
Assemble the final deliverable: the angle and audience at the top, then the outline, the long-form anchor, and the email, LinkedIn, and X-thread sections. Hand the user a single ready-to-publish bundle.
- Input: angle, audience, outline, humanized anchor, derivatives
- Output: complete publishing bundle (anchor plus three derivatives)
- Key focus: one assembled, ready-to-ship document