jtbd

name: jtbd-map version: 1.0.0 description: | Synthesizes patterns into a structured Job Map (YAML + Markdown). Accepts a patterns file (.jtbd/patterns/*.yml). Outputs to .jtbd/jobs/. Use when: "create job map", "jtbd map", "job mapping". allowed-tools:

• Bash
• Read
• Write
• Glob
• AskUserQuestion

Preamble

Check for the .jtbd/ directory. Make sure the .jtbd/jobs/ directory exists.

[ -d ".jtbd/patterns" ] || { echo "Error: .jtbd/patterns/ directory not found. Run /jtbd-patterns first."; exit 1; }
mkdir -p .jtbd/jobs/

Read Input

1. If the user provided an argument, use that file path.
2. If no argument is provided, look for the most recent file in .jtbd/patterns/. If found, ask the user to confirm using it. If not found, ask for a path.

Use the Read tool to read the patterns file.

Process Data

Instruct the LLM to deduce the chronological steps of the dominant job identified in the patterns file. For each step, identify friction points and propose high-level opportunities based on the clustered jobs, emerging signals, and timeline patterns.

Output Generation

You must generate TWO files:

1. YAML Data File: Write to .jtbd/jobs/[job-slug].yml. Format:

job: "[Main Job Statement]"
steps:
  - name: "[Step 1]"
    friction: "[Friction]"
    opportunity: "[Opportunity]"
  # ... more steps

2. Markdown Visual File: Write to .jtbd/jobs/[job-slug].md. Format as a clean document with a Markdown table for the steps:

# Job Map: [Main Job Statement]

| Step | Friction | Opportunity |
|---|---|---|
| [Step 1] | [Friction] | [Opportunity] |

Inform the user where both files were saved.