Back to skill

Security audit

github-api-push-workaround

Security checks for vulnerabilities and agentic risk

Overview

This document-only skill is purpose-aligned for Git push troubleshooting, but it can guide repository publication and private-to-public changes without strong safety gates or scoping.

Review generated commands before use, especially any git push, Git Data API ref update, or private-to-public repository change. Confirm the target remote, branch, repository visibility, and repository contents, and avoid exposing tokens or secrets in command history or published code.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (6)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill declares broad trigger phrases around 'push GitHub', publishing repositories, token failure, and push troubleshooting, which can match common developer requests and auto-invoke guidance for state-changing remote operations. Because the skill is designed to generate commands for pushing code and changing repository settings, loose activation scope increases the chance of unintended use in contexts where the user did not explicitly request publication or remote modification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This section describes generating push commands and private-to-public repository changes, both of which alter remote state and can expose source code or secrets if used incorrectly. The lack of an explicit warning about irreversible publication, branch/ref modification, and visibility changes is dangerous because users may follow generated commands without understanding the disclosure and integrity impact.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The usage section says the skill is triggered by natural-language keywords but does not define constrained boundaries, approval checks, or exclusions. In a skill that produces direct push commands and API workflows against remote repositories, unconstrained invocation increases the risk that benign troubleshooting requests are escalated into actionable instructions affecting remote code or repository configuration.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The Chinese section similarly instructs on remote push and private-to-public operations without clearly warning that these actions can disclose private code, tokens in history, or other sensitive repository contents. In context, the skill is operational and action-oriented, so omission of impact warnings materially raises the chance of accidental data exposure or unintended irreversible repository changes.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

Line L13 lists activation phrases including '发布仓库到远端' and especially '令牌失效了', which are broad natural-language expressions that could arise in many contexts beyond this specific skill. The file does not provide exclusion conditions or negative examples to narrow when the skill should or should not activate.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The file hardcodes both English and Chinese sections as the presented documentation format, but does not state that language selection is optional or user-driven. Under the stated policy rule, locale/language constraints should either offer opt-in choice or be clearly justified.

Content

No source excerpt is available for this finding.