Back to skill

Security audit

validate-pr

Security checks for vulnerabilities and agentic risk

Overview

The skill is marked suspicious because scanners report a high-confidence risk of unsafe input handling and a destructive test that may proceed despite a dirty-worktree warning; the scan was partial.

Consider avoiding installation until the unsafe filename handling and revert-test safety check are reviewed and corrected. Treat the evidence as incomplete: the Skillspector scan is partial because several referenced paths could not be resolved, and its overall analysis reports incomplete coverage.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md
Finding

PR-controlled YAML filenames are interpolated into executable Python

Content
View full analysis

Impact

An attacker can execute commands with the validation process's permissions, potentially accessing local credentials or modifying files.

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (3)

The promised dirty-worktree safety abort is not implemented before destructive revert testing.

High
Category
Excessive Agency
Confidence
99% confidence
Finding

The shell check prints a skip message but does not prevent execution from reaching the destructive revert-test steps; it also misses staged changes and untracked files. Although the prose requests an abort, the executable example does not enforce it, so subsequent edits and git checkout -- <file> can discard pre-existing unstaged work.

Content

No source excerpt is available for this finding.

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
70% confidence
Finding

Shadow Command Trigger: 'diff' conflicts with built-in command 'diff'

Content

No source excerpt is available for this finding.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
70% confidence
Finding

Overly Broad Trigger: 'pr' is too short and may match unintended inputs

Content

No source excerpt is available for this finding.