pm-jtbd-synthesislisted

Synthesize Jobs-to-be-Done from one or more customer interview transcripts. Extracts Forces of Progress (push, pull, anxiety, habit) with verbatim quote + line-number attribution, clusters jobs across transcripts, and emits three artifacts — a full synthesis report, job stories in canonical When/I-want/So-I-can format, and an Opportunity-Solution Tree stub. Triggers on "synthesize calls", "synthesize interviews", "jtbd", "jobs to be done", "discovery synthesis", "what did customers say", "interview synthesis", "forces of progress", or when the user points at a folder of call transcripts and asks for product insight. Composes downstream of `transcribe` (audio → text, multiplai-media plugin) and `extract-insights` (general extraction, multiplai-research plugin); composes upstream of `pm-persona-codifier`.
spikelab/multiplai-cc-mktplace · ★ 0 · AI & Automation · score 69

Install: claude install-skill spikelab/multiplai-cc-mktplace

# pm-jtbd-synthesis Turn raw customer-interview transcripts into discovery artifacts a product team can act on. This is **synthesis**, not summary — the output is structured around the four Forces of Progress (Bob Moesta) and clustered jobs, not around what each speaker said in order. ## Arguments | Arg | Description | Default | |-----|-------------|---------| | **source** | File path, folder path, or comma-separated list of transcript files | *(required)* | | `--depth` | `quick` (forces + top jobs only) or `full` (forces + jobs + OST stub) | `full` | | `--audio` | `true` if any inputs are audio files (will call `transcribe` first) | `false` | If the user gives a folder, glob all `*.txt`, `*.md`, `*.vtt`, `*.srt` files inside it. Confirm the file list with the user before processing if there are more than 5 files. ## Input Handling 1. If `--audio true`, route each audio file through the `transcribe` skill first (ships in the **multiplai-media** plugin — if it isn't installed, ask the user to provide text transcripts); save transcripts to a sibling folder; then proceed. 2. Read each transcript with the Read tool so you get line-numbered output. Line numbers are mandatory anchors in the output — without them the synthesis is un-verifiable. 3. If a transcript is over ~2000 lines, read it in chunks and track the line offset. 4. If there are multiple speakers, identify which speaker is the customer/interviewee. Apply forces extraction only to customer utterances. Interviewer