CodeAlive Context Engine

• Outcome: Pass
• Latest audit: May 29, 2026 · 8:22 PM UTC
• Version: 1.0.0

Overview

This is a disclosed CodeAlive integration for remote code search and Q&A, with credential and privacy caveats but no hidden or destructive behavior found.

Install only if you trust CodeAlive and the skill publisher with the repositories available to your CodeAlive API key. Prefer the interactive setup over passing keys in chat or command lines, verify CODEALIVE_BASE_URL points to the intended service, use a scoped key where possible, and avoid prompts or artifact fetches that would expose secrets, regulated data, or proprietary code you are not allowed to send to the service.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• 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 (8)

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

)
            return r.returncode == 0
        elif system == "Windows":
            r = subprocess.run(
                ["cmdkey", f"/generic:{SERVICE_NAME}", "/user:codealive", f"/pass:{api_key}"],
                capture_output=True, text=True, timeout=10,
            )

• Confidence: 87% confidence
• Finding: r = subprocess.run( ["cmdkey", f"/generic:{SERVICE_NAME}", "/user:codealive", f"/pass:{api_key}"], capture_output=True, text=True, timeout=10, )

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The skill states that chat/search functionality sends questions to a remote AI service with full indexed codebase context, but it does not provide a prominent privacy warning about sensitive source code, internal architecture, or proprietary data being transmitted off-host. In a code-intelligence skill, this context increases risk because users may query confidential repositories and assume the operation is local or low-risk.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The skill explicitly recommends passing the API key on the command line, noting only that it is visible in chat history, but it omits stronger warnings that command-line secrets may also be exposed via shell history, process listings, logging, and telemetry. This can lead to credential leakage and unauthorized use of the remote service.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: This CLI forwards user-provided queries and repository/workspace identifiers to external API calls via `client.search(...)` and `client.chat(...)` without any explicit notice, consent prompt, or sanitization boundary in the workflow. In a code-exploration skill, those inputs can reveal sensitive project names, internal architecture topics, debugging symptoms, dependency usage, or other proprietary metadata to a third-party service, creating a privacy and data-governance risk even if the feature is functioning as intended.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The chat method sends user questions and repository/workspace selections to a remote CodeAlive API, which can expose sensitive internal codebase context or proprietary project metadata if users are unaware of the external transmission. In an agent skill context, this is more dangerous because developers may assume code-assistance operations are local unless explicitly warned.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: fetch_artifacts sends artifact identifiers to an external API in order to retrieve code content, which may cause source files or symbols from private repositories to be fetched and processed remotely without a clear warning at the call site. In this skill's context, that data flow is central functionality, but it still creates a real confidentiality risk if users do not understand that repository content leaves the local environment.

Missing User Warnings

Medium

• Confidence: 84% confidence
• Finding: The script sends the supplied API key to a remote service for verification immediately after entry, but the user flow does not explicitly call out that the key will be transmitted over the network. In a setup utility that handles secrets, lack of clear disclosure can undermine informed consent and increase the chance of unintended credential exposure to alternate endpoints if CODEALIVE_BASE_URL is changed.

Missing User Warnings

Low

• Confidence: 76% confidence
• Finding: The setup flow silently checks environment variables and OS credential stores for an existing key before asking the user, without making that behavior explicit. While local-only, transparent handling of credential discovery is important in security-sensitive tooling to avoid surprising users and accidental reuse of stale or unintended credentials.

VirusTotal

66/66 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.