Read AI

Fetch and manage Read AI meeting data - summaries, transcripts, action items, and engagement metrics. Use when the user asks about meetings, meeting notes, m...

Security Scan

The skill's code, instructions, and requirements are coherent with a Read AI meeting-data integration and do not request unrelated credentials or suspicious installs.

Purpose & Capability

Name/description match the included client, search, lifecycle pull, and webhook receiver scripts; requested local API key file and local storage under ~/.readai are appropriate for a meeting ingest/search tool. Referencing Limitless (pendant) and Read AI APIs is consistent with the skill purpose.

Instruction Scope

Runtime instructions tell the agent to read an API key from ~/.config/readai/api-key, run the included Python scripts, and (optionally) run a webhook receiver. The scripts only access meeting data, local files under ~/.readai, and the Read AI / Limitless APIs. Minor scope notes: limitless_pull optionally invokes a local 'claude' CLI when --ai is used, and webhook_receiver can be bound to 0.0.0.0 if the user passes that flag (SKILL.md warns about this).

Install Mechanism

No install spec is provided (instruction-only); scripts run with the system Python and use the requests library. Nothing is downloaded from third-party URLs or written to system-wide locations by an installer.

Credentials

The skill does not request environment variables or cloud credentials; it expects a local API key file (~/.config/readai/api-key). The code reads READAI_TIMEZONE if present (optional). Requiring an API key (stored in a user file) is proportional to the stated purpose. Note the client also checks a fallback path (~/.config/limitless/api-key) which is reasonable for supporting 'Limitless' but is an additional location to be aware of.

Persistence & Privilege

always is false and the skill does not request permanent platform-level privileges. It writes meeting and log files only under the user's home (~/.readai, ~/.config/readai) and does not modify other skills or global agent configs. The webhook can be exposed to the network if the operator intentionally binds to 0.0.0.0.

Assessment

This skill appears to do what it says: it queries Read AI / Limitless APIs and saves meeting data under your home directory. Before installing, consider: 1) the API key is stored in a plain file (~/.config/readai/api-key) — protect that file (SKILL.md suggests chmod 600). 2) The Limitless pull optionally calls a local 'claude' CLI when you use the --ai flag — only enable that if you trust the 'claude' binary on your system. 3) The webhook receiver defaults to localhost but can be bound to all interfaces; only bind publicly if you know what you’re doing and secure the endpoint (use a firewall, reverse proxy with TLS, or tunnel). If those caveats are acceptable, the skill is internally consistent with its purpose.

Read AI Integration

Pull meeting intelligence from Read AI via REST API or webhook receiver.

Auth

Two data sources supported:

1. Limitless Pendant (primary) - Captures all conversations including meetings via API
2. Read AI Webhook - Real-time meeting summaries pushed from Read AI

API key stored at ~/.config/readai/api-key (same as Limitless key if using pendant).

# One-time setup
mkdir -p ~/.config/readai
echo "YOUR_API_KEY" > ~/.config/readai/api-key
chmod 600 ~/.config/readai/api-key

For Limitless: Get key from limitless.ai dashboard. For Read AI API: Get key from Read AI Dashboard > Settings > Integrations > API Keys.

Quick Commands

List Recent Meetings

python3 scripts/list_meetings.py              # Last 7 days
python3 scripts/list_meetings.py --days 30    # Last 30 days
python3 scripts/list_meetings.py --today      # Today only
python3 scripts/list_meetings.py --json       # JSON output

Get Meeting Details

python3 scripts/readai_client.py get <meeting_id>
python3 scripts/readai_client.py get <meeting_id> --transcript   # Full transcript
python3 scripts/readai_client.py get <meeting_id> --actions      # Action items only

Search Meetings

python3 scripts/search_meetings.py "quarterly review"
python3 scripts/search_meetings.py "budget" --days 30
python3 scripts/search_meetings.py "action items" --speaker "Brandon"

Export Meeting Summary

python3 scripts/readai_client.py export <meeting_id>             # Markdown
python3 scripts/readai_client.py export <meeting_id> --format json

Webhook Setup

For real-time meeting data, set up the webhook receiver:

1. Run: python3 scripts/webhook_receiver.py --port 9010
2. In Read AI Dashboard > Settings > Integrations > Webhooks
3. Add URL: http://<your-server>:9010/webhook/readai

Webhook data is stored in ~/.readai/meetings/YYYY-MM-DD/.

See references/api-reference.md for full API documentation.

Data Structure

Meetings include:

• Summary - AI-generated meeting recap
• Transcript - Full speaker-attributed transcript
• Action Items - Tasks with assignees
• Topics - Key discussion topics
• Participants - Attendees with engagement metrics
• Decisions - Key decisions made
• Duration/Timing - Start, end, duration

Local Data

Data is stored/symlinked at ~/.readai/:

~/.readai/
├── meetings/                # Webhook-received meetings
│   └── YYYY-MM-DD/
│       ├── <timestamp>_<title>.json
│       └── <timestamp>_<title>.md
├── lifelogs/                # Limitless pendant data (symlink)
│   └── YYYY-MM-DD/
│       ├── raw_lifelogs.json
│       ├── entries.json
│       └── digest.md
└── index.json               # Meeting index for search

Limitless Lifelog Pull

# Pull today's pendant data (includes meetings)
python3 scripts/limitless_pull.py --today

# Pull specific date
python3 scripts/limitless_pull.py 2026-02-19

# Pull with AI summary
python3 scripts/limitless_pull.py --today --ai