ClawHub
Back to skill

Security audit

Standup From Yesterday Commits

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent standup-summary helper, but users should review any Slack destination before posting or enabling auto-posting.

Install this if you want standup drafts from GitHub and Jira. Before connecting Slack or enabling any schedule, confirm the exact workspace, channel, audience, and message preview, and avoid auto-posting unless you intentionally want recurring updates shared there.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
85% confidence
Finding
The skill's primary purpose is summarization, but the 'What's next?' section expands it into posting to Slack and scheduling unattended auto-posts. That broadens data flow from GitHub/Jira into an external messaging system without requiring an explicit confirmation step, creating a real risk of unintended disclosure of internal work details.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation suggests posting or auto-posting the generated standup to Slack but does not warn the user that commit messages, PR titles, and Jira issue details may contain confidential engineering information. In context, the skill aggregates data from multiple internal systems, so silent forwarding to Slack meaningfully increases the chance of oversharing to the wrong audience or channel.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.