Meeting Notes Extraction

Turn a meeting transcript into notes: summary, decisions, action items with owners, key points and follow-ups.

Pay by wallet (x402) AI-generated meeting-notes-extract

Example

Input
{
  "transcript": "test_value"
}
Answer
{
  "summary": "Unable to extract meaningful meeting notes. The transcript contains only a test value with no actual meeting content, attendees, decisions, or discussion points.",
  "sentiment": "unclear",
  "word_count": 1,
  "action_items": [],
  "meeting_type": "other",
  "decisions_made": [],
  "follow_up_needed": [],
  "questions_raised": [],
  "transcript_length": 10,
  "attendees_mentioned": [],
  "key_discussion_points": []
}

Inputs

FieldTypeRequiredWhat it is
transcriptstringYesMeeting transcript text

Output fields

How reliably each field appears in a successful answer.

FieldTypeAppears
summarystringAlways (present in every successful answer)
action_itemsarrayAlways (present in every successful answer)
decisions_madearrayAlways (present in every successful answer)
key_discussion_pointsarrayAlways (present in every successful answer)

How to call it

The same tool through each route. MCP clients reach it through Strale’s server at https://api.strale.io/mcp; the call below is what the agent sends.

# Needs an API key. max_price_cents caps what this call may cost.
curl -X POST https://api.strale.io/v1/do \
  -H "Authorization: Bearer $STRALE_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "capability_slug": "meeting-notes-extract",
    "inputs": {
      "transcript": "test_value"
    },
    "max_price_cents": 20
  }'

Limitations

  • Implicit agreements and commitments can be missed

    AI extraction of action items and decisions — nuanced verbal agreements or implicit commitments may be missed

    What to do instead: Distribute extracted notes to all participants for confirmation before treating action items as final

  • Output size cap per call

    Structured output is limited to about 1,500 output tokens per call; requests exceeding this limit are refused rather than returned truncated. Requests with large results can exceed this limit, and repeating the same request does not help.

    What to do instead: Provide a shorter transcript excerpt per call

Related tools