A persona-weighted governance system that ensures AI-generated emails meet policy and safety standards before being sent.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install consensus-send-email-guard
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install consensus-send-email-guard using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Consensus Send Email Guard?
consensus-send-email-guard serves as a critical outbound communication guardrail for AI agents. It leverages a persona panel to evaluate email drafts, ensuring that high-stakes communication remains professional, compliant, and accurate. By integrating this into Openclaw Skills, developers can prevent hallucinations and policy-violating claims in external-facing messages before they are irreversibly dispatched.
This skill transforms raw AI generation into governed, auditable actions. It uses a weighted approval policy to aggregate votes based on reputation, providing a layer of human-like oversight to automated workflows. It is a production-ready solution for teams who need to maintain strict quality control over AI-driven interactions.
Consensus Send Email Guard Use Cases
- Customer-facing outbound messaging and support responses
- Partner and legal-sensitive communications requiring high accuracy
- Automated marketing campaign quality gates
- Mitigating hallucinated promises and ensuring policy-compliant outreach
How Consensus Send Email Guard Works
- The skill receives an email draft and input data via the canonical invoke entrypoint.
- It evaluates the draft using a persona panel or externally supplied votes depending on the configured mode.
- Votes are aggregated based on reputation weights to determine the final consensus.
- The system checks for hard-block categories such as sensitive data, medical/legal advice, or disallowed guarantees.
- A final decision of APPROVE, BLOCK, or REWRITE is generated in a strict JSON format.
- The decision and supporting artifacts are written to the board ledger for full auditability.
Consensus Send Email Guard Setup
To install this skill as part of your Openclaw Skills environment, use npm:
npm i consensus-send-email-guard
Ensure you have Node.js and tsx installed. You must also configure the following environment variables to manage the state ledger:
export CONSENSUS_STATE_FILE="path/to/state.json"
export CONSENSUS_STATE_ROOT="path/to/root/"
Run the guard using the following command:
node --import tsx run.js --input ./examples/email-input.json
Consensus Send Email Guard Data Schema & Taxonomy
The skill organizes data through a board ledger and state files, ensuring transparency in the decision-making process. It uses the following metadata taxonomy:
| Data Component | Description |
|---|---|
| decision | The final outcome: APPROVE, BLOCK, or REWRITE |
| artifacts | Evidence and logs of the persona evaluation process written to the ledger |
| reputation | Metadata tracking the influence and accuracy of specific evaluators over time |
| board_state | The centralized ledger tracking all outbound communication decisions and governance actions |
Consensus Send Email Guard Advanced Features
- Dual operational modes: external_agent for caller-supplied votes and persona for internal persona-set evaluation
- Idempotent retries to prevent duplicate reputation mutation and maintain ledger integrity
- Weighted approval policy that calibrates evaluator influence based on historical accuracy
- Strict JSON contracts designed for seamless integration into complex automation pipelines
- Integration with the Openclaw Skills ecosystem via standardized consensus-interact contracts
SKILL.md
Loading
Related Openclaw Skills
Consensus Persona Respawn
A persona lifecycle management tool that replaces underperforming AI evaluators with optimized successors based on historical mistake patterns.
Consensus Persona Generator
A specialized utility for generating and persisting diverse multi-agent persona sets to ensure unbiased consensus in AI workflows.
Consensus Persona Engine
A deterministic state transition layer for applying reputation updates and emitting explicit reputation delta artifacts.
Consensus Permission Escalation Guard
A deterministic safety gate for IAM and privilege elevation requests in automated agent workflows.
Cursor Agent CLI
A terminal-based AI coding assistant that provides interactive refactoring, automated reviews, and background execution via the Cursor agent.
Weekly Report Generator
An automated workflow that fetches team data, leverages AI for content synthesis, and generates professional Word documents for weekly reporting.
METADATA
Featured*
