SkillPress — 技能锻造炉

Security checks across malware telemetry and agentic risk

Overview

This looks like a skill-scaffolding helper, but its file-writing path appears under-scoped enough that users should review it before installing.

Install only if you trust the publisher and are comfortable reviewing the generated file paths. Use simple slugs with no slashes, dots, or absolute paths, and confirm where files will be written before allowing the skill to run.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
98% confidence
Finding
`slug` is used directly in `skills_dir / slug` with no validation or normalization checks, so values like `../outside` or absolute paths can escape the intended `skills` directory. Because the tool then creates directories and writes `SKILL.md` and `__init__.py`, an attacker or careless user could overwrite or plant files elsewhere on the filesystem within the process's permissions.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger phrase '把这个流程做成技能' is broad enough to overlap with ordinary conversation about process improvement, which can cause the skill to activate unexpectedly. In a skill that generates files and scaffolding, accidental activation could lead to unreviewed artifact creation or capture of more conversation context than the user intended.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal