mirror of
https://github.com/msitarzewski/agency-agents
synced 2026-04-25 11:18:05 +00:00
56 lines
2.1 KiB
Markdown
56 lines
2.1 KiB
Markdown
# Workflow Example: Book Chapter Development
|
|
|
|
> A focused single-agent workflow for turning rough source material into a strategic first-person chapter draft with explicit revision loops.
|
|
|
|
## When to Use This
|
|
|
|
Use this workflow when an author has voice notes, fragments, or strategic notes, but not yet a clean chapter draft. The goal is not generic ghostwriting. The goal is to produce a chapter that strengthens category positioning, preserves the author's voice, and exposes open editorial decisions clearly.
|
|
|
|
## Agent Used
|
|
|
|
| Agent | Role |
|
|
|-------|------|
|
|
| Book Co-Author | Converts source material into a versioned chapter draft with editorial notes and next-step questions |
|
|
|
|
## Example Activation
|
|
|
|
```text
|
|
Activate Book Co-Author.
|
|
|
|
Book goal: Build authority around practical AI adoption for Mittelstand companies.
|
|
Target audience: Owners and operational leaders of 20-200 person businesses.
|
|
Chapter topic: Why most AI projects fail before implementation starts.
|
|
Desired draft maturity: First substantial draft.
|
|
|
|
Raw material:
|
|
- Voice memo: "The real failure happens in expectation setting, not tooling."
|
|
- Notes: Leaders buy software before defining the operational bottleneck.
|
|
- Story fragment: We nearly rolled out the wrong automation in a cabinetmaking workflow because the actual problem was quoting delays, not production throughput.
|
|
- Positioning angle: Practical realism over hype.
|
|
|
|
Produce:
|
|
1. Chapter objective and strategic role in the book
|
|
2. Any clarification questions you need
|
|
3. Chapter 2 - Version 1 - ready for review
|
|
4. Editorial notes on assumptions and proof gaps
|
|
5. Specific next-step revision requests
|
|
```
|
|
|
|
## Expected Output Shape
|
|
|
|
The Book Co-Author should respond in five parts:
|
|
|
|
1. `Target Outcome`
|
|
2. `Chapter Draft`
|
|
3. `Editorial Notes`
|
|
4. `Feedback Loop`
|
|
5. `Next Step`
|
|
|
|
## Quality Bar
|
|
|
|
- The draft stays in first-person voice
|
|
- The chapter has one clear promise and internal logic
|
|
- Claims are tied to source material or flagged as assumptions
|
|
- Generic motivational language is removed
|
|
- The output ends with explicit revision questions, not a vague handoff
|