adventure-chat-ingest

Ingest recorded user conversations into a structured product discovery knowledge base.

1|Updated Jun 30, 2026
One-click install
npx skills add https://github.com/Lia-Creative/lia-plugins --skill adventure-chat-ingest-lia-creative
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: adventure-chat-ingest
Source: https://github.com/Lia-Creative/lia-plugins/tree/main/lia-tools/skills/adventure-chat-ingest
Command: npx skills add https://github.com/Lia-Creative/lia-plugins --skill adventure-chat-ingest-lia-creative

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Recorded conversations with real users contain valuable product insights, but without a consistent intake process they get lost, duplicated, or disconnected from the people and problems they describe. This Skill turns one recorded conversation (URL plus transcript) into durable, sourced product context across a discovery knowledge base. ## Core Features & Use Cases - Transcript tidying and chat notes: Cleans up timestamped .srt transcripts (fixing known mis-hearings) and produces a structured chat note with verbatim quotes and typed extractions of problems, scenarios, and insights. - Adventurer profiles and problem tracking: Creates or updates a profile per person interviewed, and feeds every problem heard into a problem register with evidence rows and timestamps. - Scenario and insight capture: Updates workflow scenarios and an insights ledger with confidence levels, plus full housekeeping across indexes, registers, and logs. - Use Case: After recording an interview with a user about how they work, hand the recording URL and transcript to this Skill to produce the chat note, update their profile, log new problems, and propose actionable tickets. ## Quick Start Ingest this adventure chat using the recording URL and transcript I have provided, and update the discovery layer accordingly.

Frequently Asked Questions about adventure-chat-ingest

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I process a user interview transcript into product insights?▼

Provide the recording URL and a timestamped .srt transcript. The Skill tidies the transcript, writes a structured chat note with verbatim quotes, updates the person's profile, and feeds problems, scenarios, and insight candidates into the discovery layer.

What transcript format is required for interview ingestion?▼

A timestamped .srt file is the required format. A plain Markdown transcript prompts a request for the .srt, and with no transcript at all the process stops, since a chat note cannot be built from memory.

How is this different from product feedback ingestion?▼

This Skill handles person-first conversations about who someone is and how they work. Recordings of product walkthroughs, bugs, or feature feedback route to the companion toy-feedback-ingest skill instead, with cross-linking when a chat mixes both.

Does it create tickets or bug reports automatically?▼

No. Nothing is created automatically in Linear. The Skill presents a proposed tick-list for approval, and bugs about a product heard mid-chat follow the separate feedback-ingest bug convention.

What happens when a second interview with the same person arrives?▼

The existing adventurer profile is updated rather than duplicated. Contradictions are resolved openly by changing the claim and noting the change in the chat-log row, never leaving two versions standing.

How are off-the-record requests handled in interview notes?▼

Anything a person asks to keep off the record stays off everywhere, including the transcript if requested mid-recording. The note records that a redaction exists without recording what was redacted.