ClawHub
Back to skill

Security audit

planning-toolkit

Security checks across malware telemetry and agentic risk

Overview

This skill is a planning and iteration toolkit that writes local planning files and bounded loop state, with no evidence of hidden data access, exfiltration, or destructive behavior.

Install only if you want a workflow that creates persistent planning artifacts in your repository and may guide agents through bounded iterative work. Use simple task slugs, review generated plans before execution, and avoid enabling loop or cleanup modes unless you have clear stopping criteria and are comfortable with local file changes.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill clearly instructs the agent to read and write files (for example creating `.planning/...`, copying templates, and updating planning artifacts), yet no explicit permissions are declared. This creates a capability/permission mismatch that can cause unsafe or unexpected filesystem modification when the skill is invoked, especially because it encourages persistent state writes across many phases.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The iterative execution trigger list is overly broad and includes common natural-language phrases such as iterative improvement, auto-retry, or performance iteration. That makes accidental activation plausible during ordinary user requests, which could push the agent into repeated loops, extra tool use, unnecessary file writes, or prolonged autonomous behavior without clear user intent.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
sed -i.bak "s/\[YYYY-MM-DD HH:MM\]/$TIMESTAMP/g" "$PLAN_DIR/task_plan.md"
sed -i.bak "s/\[YYYY-MM-DD HH:MM\]/$TIMESTAMP/g" "$PLAN_DIR/findings.md"
sed -i.bak "s/\[YYYY-MM-DD HH:MM\]/$TIMESTAMP/g" "$PLAN_DIR/progress.md"
rm -f "$PLAN_DIR/"*.bak

# Create .active_plan pointer
echo "$PLAN_DIR" > .planning/.active_plan
Confidence
84% confidence
Finding
rm -f "$PLAN_DIR/"

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.