From ravn-ai-toolkit
Converts raw meeting transcript .txt files into structured .md notes with metadata, TL;DR, key topics, action items, and quotes. Useful for processing meeting recordings or chat logs.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ravn-ai-toolkit:transcript-notesThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Convert a raw meeting transcript `.txt` file into a structured `.md` notes file.
Convert a raw meeting transcript .txt file into a structured .md notes file.
.txt file — supports speaker-turn format (speaker name on its own line, dialogue as paragraph) and timestamped formats (e.g., [00:01:23] Speaker: text)**bold title** -- one-line summary bullet, then a ### subsection with 2-5 sentences or bullets expanding it**Person/Group:** action description format> "quote" -- Speaker Name (context) format — prioritize quotes that capture decisions, insights, or strong positions over casual remarksnotes/{series-slug}/.md file mirroring the transcript filename (e.g., 001.txt → 001.md, kickoff-meeting.txt → kickoff-meeting.md)# {Series Name} - {Episode NNN or Date}
## Meeting Metadata
- **Series:** {series name}
- **Episode:** {NNN, if applicable}
- **Date:** {date, if inferable}
- **Duration:** {duration, if inferable}
- **Participants:** {comma-separated full names}
## TL;DR
{Single paragraph, 2-4 sentences}
## Key Topics
- **{Topic}** -- {One-line summary}
- **{Topic}** -- {One-line summary}
...
### {Topic Heading}
{2-5 sentences or bullet points}
### {Topic Heading}
...
## Action Items
- **{Person/Group}:** {Action description}
...
## Quotes
> "{Quote}" -- {Speaker Name} ({optional context})
...
### subsections must correspond 1:1 and appear in the same order-- (double dash) not — (em dash) for separators in topic summaries and quotesUser: "Process the meeting transcript at transcripts/ai-platform-sync/003.txt — generate formatted notes with metadata, key topics, action items, and quotes"
Expected behavior: Reads the transcript file, extracts participants, topics, action items, and quotes, then writes a structured .md notes file.
User: "Create a template agenda for our weekly sync meeting"
Expected behavior: This is meeting agenda creation, not transcript processing. The user wants to plan a future meeting, not convert an existing recording.
Error: Cannot determine series name
Error: No clear action items found
Error: Cannot estimate duration
Error: Transcript has overlapping or unclear speaker labels
Error: Transcript has no speaker labels
Error: Transcript is in a non-English language
Error: Multiple transcript files for one meeting (Part 1, Part 2)
npx claudepluginhub ravnhq/ai-toolkitProcesses raw meeting transcripts into structured Obsidian notes with YAML frontmatter, action items, summary, and formatted transcript. Useful for Granola transcripts or direct pastes.
Summarizes meeting transcripts into structured Markdown notes with date, participants, topic, decisions, points, action items, and open questions.
Processes audio recordings, transcripts, podcasts, lectures into structured Obsidian notes with action items, decisions, glossary via intake interview. Multilingual support.