Security audit

Hoist

Security checks for vulnerabilities and agentic risk

Overview

Hoist is a coherent infrastructure-management skill, but users should treat its environment-variable and database outputs as sensitive because they can contain real secrets.

Install only if you want the agent to help administer real infrastructure. Use least-privilege provider tokens, confirm destructive or public-exposure actions carefully, and avoid asking the agent to print or summarize env exports, database connection strings, or logs that may contain secrets.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill explicitly states that `hoist env list` returns real values in JSON mode and also documents `env export`, which can expose secrets such as API keys, database passwords, and tokens to the agent output channel. In an agent context, this is dangerous because secrets may be surfaced to logs, tool transcripts, downstream models, or users without any warning or masking guidance.

Static analysis

No suspicious patterns detected.