Encrytion Decryption Toolkit

• Outcome: Review
• Latest audit: Jul 9, 2026 · 3:55 PM UTC
• Version: 1.0.1

Overview

This skill is a legitimate remote cryptography toolkit, but it can send plaintext, encryption keys, HMAC secrets, and private signing keys to AgentPMT without enough safeguards or warnings.

Install only if you intentionally want AgentPMT to perform cryptographic operations remotely. Do not submit long-lived private keys, production HMAC secrets, decryption keys, access tokens, or sensitive plaintext unless you trust AgentPMT's retention, logging, and access controls; prefer local cryptography for durable secrets.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The activation keywords are overly broad and include generic terms like 'decrypt' and 'value', which can cause the skill to be selected in contexts unrelated to this specific remote cryptography service. Because this skill sends data to AgentPMT-hosted endpoints, accidental activation can route sensitive plaintext, keys, tokens, or payloads to an external service when a local or safer alternative would have been more appropriate.

Missing User Warnings

High

• Confidence: 92% confidence
• Finding: The schema invites users to submit HMAC secrets and PEM-encoded private keys to a remote tool without any warning about handling highly sensitive credential material. In an AgentPMT-hosted remote call context, this increases the risk that users or upstream agents will transmit reusable secrets to external infrastructure, where they may be logged, retained, or exposed through telemetry and debugging workflows.

External Transmission

Medium

• Category: Data Exfiltration

Content

- What AgentPMT is: ../what-is-agentpmt (ClawHub: `what-is-agentpmt`, page: https://clawhub.ai/agentpmt/what-is-agentpmt; skills.sh: `npx skills add AgentPMT/agent-skills --skill what-is-agentpmt`)
- AgentPMT account MCP/REST setup: ../agentpmt-account-mcp-rest-api-setup (ClawHub: `agentpmt-account-mcp-rest-api-setup`, page: https://clawhub.ai/agentpmt/agentpmt-account-mcp-rest-api-setup; skills.sh: `npx skills add AgentPMT/agent-skills --skill agentpmt-account-mcp-rest-api-setup`)
- Marketplace product: https://www.agentpmt.com/marketplace/encrytion-decryption-toolkit
- AgentPMT main MCP server: https://api.agentpmt.com/mcp/
- AgentPMT REST invoke endpoint: https://api.agentpmt.com/products/purchase

• Confidence: 94% confidence
• Finding: This skill is explicitly designed to invoke remote AgentPMT services, meaning cryptographic inputs may be transmitted off-host to https://api.agentpmt.com/. In context, those inputs can include highly sensitive material such as plaintext secrets, HMAC secrets, private signing keys, ciphertext, and decryption parameters, so external transmission materially increases confidentiality and key-exposure risk.

External Transmission

Medium

• Category: Data Exfiltration

Content

- AgentPMT account MCP/REST setup: ../agentpmt-account-mcp-rest-api-setup (ClawHub: `agentpmt-account-mcp-rest-api-setup`, page: https://clawhub.ai/agentpmt/agentpmt-account-mcp-rest-api-setup; skills.sh: `npx skills add AgentPMT/agent-skills --skill agentpmt-account-mcp-rest-api-setup`)
- Marketplace product: https://www.agentpmt.com/marketplace/encrytion-decryption-toolkit
- AgentPMT main MCP server: https://api.agentpmt.com/mcp/
- AgentPMT REST invoke endpoint: https://api.agentpmt.com/products/purchase

• Confidence: 94% confidence
• Finding: The REST invoke endpoint also routes requests to an external AgentPMT service, creating the same exposure channel for sensitive cryptographic inputs and outputs. In a crypto toolkit, this context makes the issue more dangerous than ordinary API use because users may provide raw keys, access tokens, secrets, or plaintext expecting only transformation rather than third-party handling.

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.