lark-workflow-meeting-summary

Meeting Minutes Summary Workflow

CRITICAL, Before starting, you MUST use the Read tool to read ../lark-shared/SKILL.md, which contains information on authentication and permission handling. Then read ../lark-vc/SKILL.md to understand operations related to meeting minutes.

CRITICAL, Before starting, you MUST use the Read tool to read ../lark-vc/references/vc-domain-boundaries.md. Failure to read this will lead to errors in command usage, meeting artifact decision-making, and domain boundary responsibility judgment:

1. Understand the relationship and division of responsibilities between Calendar & VC, and Meeting Artifacts & Documents.
2. Understand the relationship between meeting artifacts (transcripts and minutes), for example: the generation conditions for transcripts and minutes are independent.
3. Understand the components of different meeting artifacts to decide which artifact data to use based on requirements.
4. Understand the standard process for meeting summarization, analysis, and information extraction.

Applicable Scenarios

• "Help me organize this week's meeting minutes" / "Summarize recent meetings" / "Generate a weekly meeting report"
• "See what meetings were held today" / "Review what meetings were held over the past week"

Prerequisites

Only user identity is supported. Ensure authorization is granted before execution:

lark-cli auth login --domain vc        # Basic (query + minutes)
lark-cli auth login --domain vc,drive   # Includes reading minute document content, generating documents

Workflow

{Time Range} ─► vc +search ──► Meeting List (meeting_ids)
                   │
                   ▼
               vc +notes ──► Minute Document tokens
                   │
                   ▼
               drive metas batch_query Minute metadata
                   │
                   ▼
               Structured Report

Step 1: Determine Time Range

Default is past 7 days. Inference rules: "today" -> current day, "this week" -> Monday of this week to now, "last week" -> Monday to Sunday of last week, "this month" -> 1st to now.

Note: Date conversion must call system commands (such as date), do not calculate mentally. Time range parameters must be formatted according to actual CLI requirements (usually YYYY-MM-DD or ISO 8601).

Step 2: Query Meeting Records

# page-size maximum is 30
lark-cli vc +search --start "<YYYY-MM-DD>" --end "<YYYY-MM-DD>" --format json --page-size 30

• Time range splitting: The maximum search time range is 1 month. To search for a longer time range, it must be split into multiple queries of one month each.
• --end is the date including the current day (i.e., when checking "today", fill both start and end with today).
• --format json outputs in JSON format, which you are better at parsing.
• --page-size 30 maximum 30 items per page.
• When page_token exists, you must continue to paginate and collect all id fields (meeting-id).

Step 3: Obtain Minute Metadata

1. Query meeting-related minute information

lark-cli vc +notes --meeting-ids "id1,id2,...,idN"   

• Query meeting minutes based on the meeting-id collected in the previous step.
• A maximum of 50 minute records can be queried at once; if more than 50, batch calls are required.
• Some meetings return no notes available, mark as "No minutes" in the final output.
• Record the note_id (minute ID), note_display_type (display type: unknown / normal / unified), note_doc_token (minute document token), and verbatim_doc_token (verbatim transcript document token) for each meeting.

Verbatim transcript routing is determined by note_display_type (see the Note domain in vc-domain-boundaries.md):

• normal: The verbatim transcript is an independent document; links/content use verbatim_doc_token.
• unified: The verbatim transcript is not an independent document and has no shareable verbatim transcript document link; when verbatim content is needed, use note +transcript --note-id <note_id> (lark-note) to pull it locally, and mark as "unified minutes" in the report.

2. Obtain links for minute documents and verbatim transcript documents

# Learn command usage
lark-cli schema drive.metas.batch_query

# Batch obtain links for minute documents and verbatim transcripts: maximum 10 documents per query
# Only query links for note_doc_token and verbatim_doc_token of normal minutes
lark-cli drive metas batch_query --data '{"request_docs": [{"doc_type": "docx", "doc_token": "<doc_token>"}], "with_url": true}'

Step 4: Organize Minute Report

Choose output format based on time span:

• Single-day summary ("today"/"yesterday"): Use "Today's Meeting Overview" as the title, list meeting time, topic, minute link, and verbatim transcript link for each meeting (unified minutes have no verbatim transcript link, mark as "unified minutes, verbatim transcript requires note +transcript to pull").
• Multi-day/Weekly report ("this week"/"past 7 days", etc.): Use "Meeting Minutes Weekly Report" as the title, including overview statistics and details for each meeting.

Step 5: Generate Document (Optional, when requested by user)

Read ../lark-doc/SKILL.md to learn cloud document skills.

lark-cli docs +create --api-version v2 --doc-format markdown --content $'<title>Meeting Minutes Summary (<start> - <end>)</title>\n<content>'
# Or append to an existing document
lark-cli docs +update --api-version v2 --doc "<url_or_token>" --command append --doc-format markdown --content $'<content>'

References

• lark-shared, Authentication, permissions (Must read)
• lark-vc, Detailed usage of +search, +notes
• lark-note, note +detail, note +transcript (unified minutes verbatim transcript)
• lark-doc, Detailed usage of +fetch, +create, +update