ClawHub
Back to skill

Security audit

GitHub App

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent GitHub App authentication helper, but users should treat its tokens and local auth changes carefully.

Install only if you intend this machine or agent to authenticate to GitHub through a GitHub App. Treat printed tokens and private keys as secrets, use separate profiles for separate clients or organizations, and use auth reset when you no longer want ghapp controlling local git/gh authentication.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
86% confidence
Finding
This markdown file describes security-relevant behaviors including printing an installation token, configuring git/gh authentication, and resetting auth state. While the commands are central to the skill's purpose, the description does not warn users that these actions affect local auth configuration and may expose sensitive credentials if tokens are printed or cached.

Self-Modification

High
Category
Rogue Agent
Content
- `ghapp config set`, `ghapp config get [key]`, `ghapp config path` — manage config
- `ghapp profile list` — list configured profiles
- `ghapp profile current` — print the profile in effect for this invocation
- `ghapp update` — self-update to latest release
- `ghapp version` — print version

gh auth modes (passed to `auth configure`)
Confidence
90% confidence
Finding
Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Static analysis

No suspicious patterns detected.