ClawHub
Back to skill

Security audit

gitee

Security checks for vulnerabilities and agentic risk

Overview

This Gitee skill is a straightforward command reference for Gitee API and git workflows, with expected but powerful repository and token usage.

Install this only if you intend to let the agent work with Gitee repositories. Use a least-privilege Gitee token, prefer read-only permissions for inspection tasks, and require explicit approval before creating issues or pull requests, updating files, or pushing branches.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Credential Access

High
Category
Privilege Escalation
Confidence
79% confidence
Finding

The skill requires a personal access token with repository permissions and repeatedly demonstrates passing it as an access_token, including in request parameters and JSON bodies. In practice, this increases the likelihood of credential exposure through shell history, process inspection, logs, proxies, or accidental debugging output, especially in automated agent environments.

Content

Scanner excerpt · SKILL.md (reported line 38)May include surrounding context.

Setup

  1. Create a Gitee personal access token with the repository permissions you need.
  2. Export the API base URL and token:
bash

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 92)May include surrounding context.

bash
PR_NUMBER=12

curl -fsS --get "$GITEE_API/repos/$OWNER/$REPO/pulls/$PR_NUMBER" \
  --data-urlencode "access_token=$GITEE_ACCESS_TOKEN"

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 134)May include surrounding context.

md
TITLE="Bug: unexpected failure"
BODY="Steps to reproduce..."

curl -fsS -X POST "$GITEE_API/repos/$OWNER/issues" \
  -H "Content-Type: application/json" \
  -d "$(jq -nc \
    --arg access_token "$GITEE_ACCESS_TOKEN" \

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill metadata and description frame repository file access as read-oriented, but the body explicitly instructs users to create or update repository contents. That mismatch can cause an agent or operator to invoke the skill in situations assumed to be non-mutating, leading to unintended remote state changes in a live repository.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill includes write-capable API calls and git push operations, but does not prominently warn that these commands modify remote repository state. In an agent context, this increases the risk of accidental issue creation, PR creation, file modification, or branch updates when the user may have expected inspection-only behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.