← Back to Skills Marketplace
Openclaw
by
johnjoubert
· GitHub ↗
· v1.0.5
· MIT-0
261
Downloads
1
Stars
0
Active Installs
8
Versions
Install in OpenClaw
/install robotomail
Description
Give an AI agent a real inbox and outbound email using Robotomail. Use this skill when the user asks you to send email, check inbox, read messages, reply to...
Usage Guidance
This skill appears to be what it says: an API-first Robotomail integration that only needs a Robotomail API key. Before installing, consider: (1) provenance — the publisher and homepage are unknown and the registry name mismatches the SKILL.md (robotomail vs Openclaw), so verify the source if you need stronger trust; (2) privilege — anyone with the provided ROBOTOMAIL_API_KEY can send and read mail for that account, so prefer creating a scoped API key limited to specific mailboxes or minimal scopes; (3) webhooks — if you configure webhooks, save the webhook secret securely (examples reference WEBHOOK_SECRET but the skill doesn't declare it); (4) data sensitivity — the agent will be able to read emails and attachments (PII), so only enable this skill for agents you trust and consider auditing outbound messages and rate limits; (5) review Robotomail's privacy and terms at robotomail.com before providing credentials. If you need higher assurance, ask the publisher for a verifiable homepage or source repository and prefer scoped keys instead of a full-account key.
Capability Analysis
Type: OpenClaw Skill
Name: robotomail
Version: 1.0.5
The robotomail skill bundle provides a legitimate interface for an AI agent to interact with the Robotomail email infrastructure API. The documentation (SKILL.md and api-reference.md) clearly outlines standard email operations such as sending, receiving, and managing mailboxes, while the included code snippets in webhook-verification.md demonstrate secure practices for signature validation. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Tags
Capability Assessment
Purpose & Capability
The SKILL.md describes an email integration (sending, receiving, webhooks, SSE, domains) and the only required credential is ROBOTOMAIL_API_KEY, which is appropriate for the stated purpose. Minor inconsistency: registry metadata lists the skill name as "Openclaw" while the SKILL.md and slug use "robotomail"; the source/homepage are unknown which reduces provenance/trust but does not contradict functionality.
Instruction Scope
Runtime instructions are limited to calling Robotomail HTTP endpoints (list mailboxes, send/read messages, manage domains/webhooks, SSE). They do not instruct reading unrelated system files or unrelated environment variables. The webhook verification examples reference storing a webhook secret in an environment variable (e.g., WEBHOOK_SECRET) but that variable is not declared as required by the skill; this is a documentation/example mismatch rather than explicit exfiltration.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to be written/executed locally, which is the lowest-risk install model.
Credentials
The single required env var (ROBOTOMAIL_API_KEY) is proportional to an API-based email integration. No unrelated credentials or filesystem config paths are requested. Note: webhook examples mention a WEBHOOK_SECRET (used to verify inbound webhooks) but that is an implementation detail and not declared as required by the skill.
Persistence & Privilege
The skill does not request always:true, does not modify other skills or system settings, and has no declared persistence requirements. Autonomous invocation is enabled (default) which is normal for skills; consider that an agent with this key can send/receive email on behalf of the account.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install robotomail - After installation, invoke the skill by name or use
/robotomail - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.5
Inbound attachments support documented. SKILL.md and api-reference.md updated to teach agents how to read attachments[] from message.received SSE/webhook payloads, fetch download_url directly, and rewrite cid: inline images in body_html. New decision-tree section for reading inbound attachments. GET /v1/attachments/{id} added to quick-reference. Two-surface contract spelled out (webhook/SSE includes download_url, REST list/get is metadata-only). Corrected SSE concurrency limit from 10 to 5.
v1.0.4
Fix display name to Robotomail
v1.0.3
Add abuse prevention docs: bounce/complaint thresholds, velocity limits, suspension response format
v1.0.2
Version consistency fix
v1.1.1
Fix version consistency in frontmatter
v1.1.0
Improved positioning, added use-case examples, SSE guidance, and Webhooks vs SSE decision tree
v1.0.1
Fix monthly send limit: paid tier is 5,000/month, not 30,000
v1.0.0
Initial release — full API reference for AI agents to send/receive email via Robotomail
Metadata
Frequently Asked Questions
What is Openclaw?
Give an AI agent a real inbox and outbound email using Robotomail. Use this skill when the user asks you to send email, check inbox, read messages, reply to... It is an AI Agent Skill for Claude Code / OpenClaw, with 261 downloads so far.
How do I install Openclaw?
Run "/install robotomail" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Openclaw free?
Yes, Openclaw is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Openclaw support?
Openclaw is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Openclaw?
It is built and maintained by johnjoubert (@johnjoubert); the current version is v1.0.5.
More Skills