generated: '2026-08-05'
method: searched
source: https://www.read.ai/post/read-ai-is-now-a-claude-connector
description: >-
  Agent Skills PUBLISHED BY READ AI, not generated by API Evangelist. Read AI ships
  three downloadable .skill bundles alongside its Claude connector; each SKILL.md is
  saved here VERBATIM from the provider's own S3 distribution. Every skill names the
  Read AI MCP server as a required connector and calls its real tools
  (list_meetings, get_meeting_by_id) — which is where this repo's MCP tool names and
  parameter names come from.
distribution:
  host: readai-assets-production.s3.us-east-1.amazonaws.com
  path: /readai_skills/
  format: .skill (zip bundle — SKILL.md + references/ + assets/)
  announced: '2026-05-27'
skills:
  - file: read-ai-customer-feedback-digest.md
    name: customer-feedback-digest
    title: Customer Feedback Digest
    summary: >-
      Weekly product-feedback rollup. Pulls discovery/demo meetings from the Read AI
      connector for a window, extracts feedback into four categories, optionally tags
      persona and vertical, tracks an 8-week rolling trend, and publishes a markdown
      snapshot plus a wiki page (Confluence/Notion/SharePoint/OneNote) and a chat
      digest (Slack/Teams).
    api: mcp/read-ai-mcp.yml
    operations: [list_meetings, get_meeting_by_id]
    required_connectors: [Read AI MCP]
    optional_connectors: [Confluence, Notion, SharePoint, OneNote, Slack, Microsoft Teams]
    source: https://readai-assets-production.s3.us-east-1.amazonaws.com/readai_skills/customer-feedback-digest.skill
    bundle_files: 7
  - file: read-ai-friday-followups.md
    name: friday-followups
    title: Friday Followups
    summary: >-
      Scheduled weekly planner. Pulls the past week's meetings and action items from
      the Read AI connector plus DMs and @-mentions from Slack or Teams, LLM-judges
      what is actually the user's to do, estimates durations in bounded buckets,
      groups by topic, and books tentative private calendar blocks on Google Calendar
      or Outlook with a daily cap and a one-command undo.
    api: mcp/read-ai-mcp.yml
    operations: [list_meetings, get_meeting_by_id]
    required_connectors: [Read AI MCP]
    optional_connectors: [Slack, Microsoft Teams, Google Calendar, Outlook Calendar]
    source: https://readai-assets-production.s3.us-east-1.amazonaws.com/readai_skills/friday-followups.skill
    bundle_files: 5
  - file: read-ai-sales-call-review.md
    name: sales-call-review
    title: Sales Call Review
    summary: >-
      Weekly coaching and deal-health report. Pulls customer-facing calls from the
      Read AI connector, identifies reps by internal email domain or CRM user list,
      then runs deal-health analysis per account and coaching analysis per rep using
      Read AI's metrics payload for talk-time, and emits one self-contained
      interactive HTML report with inline Chart.js visualizations.
    api: mcp/read-ai-mcp.yml
    operations: [list_meetings, get_meeting_by_id]
    required_connectors: [Read AI MCP]
    optional_connectors: [HubSpot, Salesforce]
    source: https://readai-assets-production.s3.us-east-1.amazonaws.com/readai_skills/sales-call-review.skill
    bundle_files: 8
notes:
  - >-
    Only SKILL.md is mirrored here. The references/ markdown and the bundled
    chart.umd.min.js asset stay at the provider's distribution URL — this repo is a
    catalog, not a redistribution point.
  - >-
    All three skills describe first-run setup wizards, dry-run modes, run logging and
    unattended scheduled execution.
x-evidence:
  fetched: '2026-08-05'
  probes:
    - {url: 'https://readai-assets-production.s3.us-east-1.amazonaws.com/readai_skills/customer-feedback-digest.skill', http_status: 200, bytes: 33717}
    - {url: 'https://readai-assets-production.s3.us-east-1.amazonaws.com/readai_skills/friday-followups.skill', http_status: 200, bytes: 22571}
    - {url: 'https://readai-assets-production.s3.us-east-1.amazonaws.com/readai_skills/sales-call-review.skill', http_status: 200, bytes: 104642}