Hushh One PKM

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 10:08 AM UTC
• Version: 1.0.0

Overview

The skill is a disclosed consent-based bridge for accessing Hushh One personal knowledge data, with clear scope and retention limits.

Install this only if you intend to let OpenClaw work with Hushh One PKM data. Keep the developer token in the local secret/environment mechanism, use the trusted MCP bridge, confirm requested scopes are narrow, and define retention controls for any downstream system that receives plaintext personal data.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 80% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 23)May include surrounding context.

md

5. Store it locally as `HUSHH_DEVELOPER_TOKEN`.
6. Restart the OpenClaw gateway after changing environment variables.

Never ask the user to paste the token into chat. Never put a developer token in a URL, prompt, file committed to git, or tool argument unless the local OpenClaw secret/env mechanism requires it.

## Consent Flow

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 85% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

md

4. If the request is pending, poll `hushh-one__check_consent_status` with the returned `request_ref` at the returned interval.
5. After consent is granted, call `hushh-one__get_encrypted_scoped_export` with the returned `grant_ref` and the original `expected_scope`.

The user approves or denies consent in the first-party Hussh One/Kai app. Do not ask the user for an approval link, do not impersonate the user, and do not bypass consent polling.

## Scope Rules

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 80% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

md

4. If the request is pending, poll `hushh-one__check_consent_status` with the returned `request_ref` at the returned interval.
5. After consent is granted, call `hushh-one__get_encrypted_scoped_export` with the returned `grant_ref` and the original `expected_scope`.

The user approves or denies consent in the first-party Hussh One/Kai app. Do not ask the user for an approval link, do not impersonate the user, and do not bypass consent polling.

## Scope Rules