Objection Source Diagnoser

Security checks across malware telemetry and agentic risk

Overview

This skill appears to generate a sales-planning markdown report, with only a low-risk file-creation side effect to be aware of.

Before installing, expect the skill to create a markdown report in the current workspace. Check the filename if the deal name contains sensitive text or unusual characters, and avoid overwriting an existing report with the same name.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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 (2)

Missing User Warnings

Low

Confidence: 95% confidence
Finding: The skill declares that it will generate and write `objection-prevention-plan-{deal}.md` into the workspace, but the user-facing description does not clearly warn that the skill modifies files. This can lead to unexpected workspace changes, accidental overwrites if naming collides, and reduced user awareness/consent around side effects, even though the file written is a predictable analysis artifact rather than arbitrary code or command execution.

Missing User Warnings

Low

Confidence: 97% confidence
Finding: The process instructions explicitly tell the agent to write `objection-prevention-plan-{deal}.md`, but there is no nearby explicit warning or consent step informing the user that a new file will be created. While the content is limited to sales-analysis output and the context is offline document processing, silent file creation is still a real workspace-integrity issue because it introduces side effects without clear disclosure.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal