Skip to content

Clawdbot skill for processing client call transcripts into actionable ClickUp tasks

Notifications You must be signed in to change notification settings

Akera-Agency/call-processor-skill

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Call Processor

A Clawdbot skill for processing client call transcripts into actionable ClickUp tasks.

What It Does

  1. Pulls transcripts from Fathom (call recording tool)
  2. Extracts actionable items — features, bugs, enhancements, action items
  3. Deduplicates against ClickUp — checks for existing tasks before creating new ones
  4. Confirms with user — shows full ticket preview, asks for approval
  5. Creates tasks — with proper notation matching your project's existing style

Installation

Via ClawdHub

clawdhub install call-processor

Manual

Clone to your skills directory:

git clone https://github.com/YOUR_ORG/call-processor.git ~/clawd/skills/call-processor

Setup

1. Configure Project Mapping

Edit references/projects.json with your project aliases:

{
  "PROJECT_NAME": {
    "aliases": ["alias1", "alias2"],
    "contacts": ["Contact Name"]
  }
}

2. Set Up ClickUp Skill

This skill depends on the ClickUp skill being installed and configured.

3. Set Up Fathom Skill

This skill uses the Fathom skill to pull transcripts.

Usage

"Process the MHP Pros call from today"
"Process yesterday's standup with The Key team"

The agent will:

  1. Find the call in Fathom
  2. Pull and analyze the transcript
  3. For each extracted item:
    • Search ClickUp for existing similar tasks
    • Show you the match (if found) with options
    • Show full ticket preview
    • Wait for your approval (Y/N/Comment/Update/Skip)
  4. Create approved tasks with correct notation

Directory Structure

call-processor/
├── SKILL.md              # Main skill instructions (loaded by Clawdbot)
├── README.md             # This file
├── .gitignore            # Excludes private data
├── scripts/              # Utility scripts
│   ├── context.py        # Get project context
│   ├── tickets.py        # List/search ClickUp tickets
│   └── roadmap.py        # Get feature roadmap
├── references/           # Reference data (customize for your org)
│   └── projects.json     # Project aliases and contacts
├── templates/            # Output templates (optional)
├── cache/                # [GITIGNORED] Cached ClickUp structure
└── calls/                # [GITIGNORED] Call transcripts & extractions

What's Gitignored (Private)

These directories contain client-specific data and should NOT be committed:

  • cache/ — Contains your ClickUp workspace IDs and list mappings
  • calls/ — Contains actual call transcripts and extracted items with client discussions

What's Shared (Public)

These can be safely shared with your team:

  • SKILL.md — The workflow instructions
  • scripts/ — Utility scripts (no hardcoded IDs)
  • references/projects.json — Template with placeholder values
  • templates/ — Output templates

Dependencies

License

MIT

About

Clawdbot skill for processing client call transcripts into actionable ClickUp tasks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages