Hire

Security checks across malware telemetry and agentic risk

Overview

This instruction-only hiring wizard is coherent with its purpose, but it can create persistent agent files, share team memory by default, and optionally set scheduled review jobs, so users should review those choices.

Before installing or using this skill, be prepared to review the generated agent's tools, autonomy, boundaries, shared memory links, and any optional cron-based performance review schedule. The behavior is disclosed and matches the skill's purpose, but it affects persistent agent configuration.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
Low
What this means

A newly created agent may be given broader tools or autonomy than the user intended if the inferred role rules are not checked.

Why it was flagged

The skill generates rules that can define a future agent's allowed actions and autonomy. This is purpose-aligned for a hiring wizard, but users should review the generated permissions.

Skill content
AGENTS.md - Role definition, responsibilities, operational rules, what they do freely vs ask first
Recommendation

Review the summary card and generated AGENTS.md/TOOLS.md before using the new agent, especially the tools, boundaries, and 'ask first' rules.

#
ASI06: Memory and Context Poisoning
Medium
What this means

The new agent may see shared user details and team memory, and information added by one agent may affect another agent later.

Why it was flagged

The skill links the new agent to shared user and memory files by default, which can expose personal/team context and allow shared memory to influence future agent behavior.

Skill content
USER.md → `../../USER.md` ... MEMORY.md → `../../MEMORY.md` (shared team context)
Recommendation

Only keep these shared links if the new agent should have shared context; otherwise replace them with isolated files or narrower context.

#
ASI10: Rogue Agents
Medium
What this means

If enabled, scheduled review conversations may continue until the schedule is changed or stopped.

Why it was flagged

A cron job is a persistence mechanism that can cause future scheduled activity. The artifact frames it as optional and asks the user for frequency, so this is a disclosure note rather than a concern.

Skill content
Create a cron job that triggers a review conversation
Recommendation

Enable performance-review scheduling only if desired, confirm the frequency, and make sure you know how to disable the cron job later.