GitCode API

• Outcome: Review
• Latest audit: May 29, 2026 at 9:26 PM UTC
• Version:
  1.3.0

Overview

This GitCode API skill is documentation-only, but it is framed as a lookup helper while including examples and references for actions that can change repositories and grant access.

Install only if you want an agent to use broad GitCode API documentation. Use a least-privilege token, avoid pasting long-lived tokens into chat or URLs, and require explicit approval before any operation that creates issues or PRs, changes files/settings, manages webhooks, transfers or deletes repositories, or adds/removes members.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The examples include creating issues, which is a state-changing write operation despite the skill being described primarily as fetch/query oriented. This mismatch can mislead users or downstream agents into granting broader trust or permissions than expected, increasing the risk of unintended remote changes.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The pull request creation example exceeds a read/query-only framing and demonstrates a repository-modifying workflow. In an agent context, this can cause unsafe assumptions about the skill's authority and lead to unintended code changes or workflow manipulation in remote repositories.

Description-Behavior Mismatch

High

• Confidence: 98% confidence
• Finding: The member addition example performs an access-control change by adding collaborators, which is far more sensitive than simple data retrieval. Because the manifest frames the skill as fetch/query focused, users may not anticipate privilege-affecting actions, creating a substantial risk of unauthorized or accidental access grants.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The examples instruct use of an authentication token but provide no handling, storage, redaction, or scope-minimization guidance. In practice this increases the likelihood of token leakage through copy/paste, logs, screenshots, source control, or over-privileged usage.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The state-changing issue-creation example lacks a warning that it will modify remote repository data. Without this context, users or agents may run it expecting a harmless query example and accidentally create records in production repositories.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: Adding a project member changes repository access and can materially affect confidentiality, integrity, and governance. Omitting a warning about the permission impact makes this especially dangerous in an agent or automation setting where users may not realize the example grants access to another account.

VirusTotal

65/65 vendors flagged this skill as clean.