← Back to Skills Marketplace
Delivery Queue
by
iPythoning
· GitHub ↗
· v1.0.0
· MIT-0
98
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install delivery-queue
Description
Schedule and send segmented messages in timed intervals to mimic human-like delivery and optimize engagement.
README (SKILL.md)
delivery-queue — Delayed Segmented Delivery
Schedule and deliver messages in timed segments to simulate human-like sending patterns.
Use Cases
- Break long product introductions into 3-5 digestible messages
- Schedule follow-ups at optimal times (e.g., prospect's local 9 AM)
- Drip campaigns: space out nurture messages over days
- Avoid WhatsApp spam detection by pacing outbound messages
Commands
deliver:schedule— Queue a message for future deliverydeliver:list— View pending deliveriesdeliver:cancel— Cancel a scheduled deliverydeliver:flush— Send all pending messages immediately
Configuration
default_delay_ms: 3000 # Delay between segments
max_segments: 10 # Max segments per delivery
timezone: "{{timezone}}" # Owner's timezone
quiet_hours: # Don't send during these hours
start: "22:00"
end: "07:00"
Usage Example
Schedule a 3-part product intro to +1234567890:
1. Company overview (send now)
2. Product highlights (send after 5 min)
3. Pricing inquiry (send after 15 min)
How It Works
- AI composes the full message sequence
- Skill splits into segments with timing
- Each segment is queued with a delivery timestamp
- Background worker sends at scheduled times
- Failed deliveries retry up to 3 times
File Structure
delivery-queue/
├── SKILL.md # This file
└── deliver.sh # Queue management script
Usage Guidance
This package only provides a local queue manager: it writes JSON files to $HOME/.openclaw/delivery-queue and can mark entries 'sent' but contains no code to actually connect to WhatsApp, Telegram, or an email SMTP/API. Before installing, ask the author how messages are delivered (what background worker or integration is required), whether API keys or SMTP credentials are needed, and where those credentials should be stored. Ensure you are comfortable that another component (or the agent) will perform sending securely — otherwise the skill will not perform its claimed function. Also note the script calls python3, sha256sum, and realpath; ensure those binaries exist and review any external integration code that will be added to handle actual message delivery.
Capability Analysis
Type: OpenClaw Skill
Name: delivery-queue
Version: 1.0.0
The skill provides a legitimate utility for scheduling and queuing messages for segmented delivery. The implementation in `deliver.sh` demonstrates defensive programming practices, including strict input validation for channels and recipients, path traversal protection in the cancel command, and safe JSON serialization using Python to prevent injection. No indicators of malicious intent, data exfiltration, or harmful prompt injection were found.
Capability Assessment
Purpose & Capability
The name/description promise scheduling and sending to WhatsApp/Telegram/email, but the provided deliver.sh only creates/manages local JSON queue entries and can mark them as sent; there is no delivery worker, no network/SMTP/third-party API integration, and no required API keys or SMTP credentials declared. A delivery skill that actually sends messages would normally require service credentials and sending logic.
Instruction Scope
SKILL.md describes a background worker and retry behavior, but the runtime instructions and deliver.sh do not implement sending or retries — they only queue, list, cancel, flush (mark as sent) and clean local entries. The instructions do not request or document how messages get delivered, which grants broad implicit responsibility to other components or the agent without making them explicit.
Install Mechanism
No install spec (instruction-only), so the code will be placed on disk as-is. This is low-risk from an install-download perspective, but the script invokes external binaries (python3, sha256sum, realpath, date) that the package metadata did not declare as required.
Credentials
The skill declares no required env vars or credentials, yet targets external messaging services where API keys/SMTP credentials would normally be required. The script optionally uses OPENCLAW_HOME but that env var was not documented in SKILL.md's 'requires'. No sensitive data is accessed by the script itself, but absence of declared credentials is disproportionate to the stated purpose.
Persistence & Privilege
The skill does not request always:true and does not modify other skills or global configuration. It creates and manages files under OPENCLAW_HOME or $HOME/.openclaw/delivery-queue (its own data), which is expected behavior for a local queue.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install delivery-queue - After installation, invoke the skill by name or use
/delivery-queue - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial publish from b2b-sdr-agent-template v3.3.1
Metadata
Frequently Asked Questions
What is Delivery Queue?
Schedule and send segmented messages in timed intervals to mimic human-like delivery and optimize engagement. It is an AI Agent Skill for Claude Code / OpenClaw, with 98 downloads so far.
How do I install Delivery Queue?
Run "/install delivery-queue" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Delivery Queue free?
Yes, Delivery Queue is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Delivery Queue support?
Delivery Queue is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Delivery Queue?
It is built and maintained by iPythoning (@ipythoning); the current version is v1.0.0.
More Skills