golang-stay-updated

• Outcome: Review
• Latest audit: Sep 15, 2026 · 1:04 AM UTC
• Version: 1.3.0

Overview

The skill is a simple Go resource guide, but it asks for workspace-editing, command-running, network, and agent-delegation permissions it does not need.

Install only if you are comfortable with the skill enabling broader agent capabilities than its resource-guide purpose requires, or restrict its tool permissions to read-only/static use. I found no evidence of malicious code, exfiltration, persistence, or destructive behavior.

A.I.G

Vulnerability Patterns

• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs
• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (1)

T05 · Unauthorized Access and Privilege Escalation

Warning

• Location: SKILL.md:15

Finding

Excessive Tool Permissions Violate Least Privilege

Content

View full analysis

Vulnerability Details

File Location: SKILL.md:15
Vulnerability Type: T05: Unauthorized Access and Privilege Escalation
Risk Level: Medium

Vulnerable Code

yaml

allowed-tools: Read Edit Write Glob Grep Bash(go:*) Bash(golangci-lint:*) Bash(git:*) Agent WebFetch WebSearch

Technical Analysis

The skill is a static guide that recommends Go resources, communities, newsletters, blogs, videos, and developers. Its documented functionality does not require modifying files, executing shell commands, invoking Git or Go tooling, or delegating work to sub-agents.

Nevertheless, the configuration authorizes Edit, Write, multiple Bash(...) command families, and Agent. It also grants network capabilities through WebFetch and WebSearch. This unnecessarily broad authorization violates the principle of least privilege and increases the consequences of malicious or untrusted instructions entering the agent context.

The reviewed skill does not itself instruct the agent to abuse these permissions. Exploitation therefore depends on another source of attacker-controlled instructions, such as prompt injection in user input or remotely retrieved content.

Attack Path

1. The skill is loaded, activating its broad tool allowlist.
2. An attacker introduces malicious instructions through user-controlled context or content processed during the session.
3. Those instructions persuade the agent to invoke an unnecessary authorized capability, such as Edit, Write, Bash(git:*), or Agent.
4. The agent modifies workspace files, runs permitted Go or Git commands, delegates attacker-directed activity, or retrieves additional untrusted content.
5. The resulting operation affects the workspace or expands the attack surface beyond the skill's legitimate informational purpose.

Impact Assessment

Successful exploitation could permit modification of files available to the agent, execution of authorized Go, GolangCI-Lint, or Git command famil ...[truncated 343 chars]

Remediation

View remediation

Remediation Suggestions

1. Remove Edit and Write, because the skill only presents informational content.
2. Remove Bash(go:*), Bash(golangci-lint:*), and Bash(git:*); none are needed for the documented behavior.
3. Remove Agent to prevent unnecessary delegation and capability expansion.
4. Remove WebFetch and WebSearch if the bundled static resource list is authoritative.
5. If current information must be retrieved, retain only the minimum network capability and require explicit user approval before accessing external content.
6. Prefer an empty tool allowlist for normal use of this static guide.
7. Add automated validation that rejects tool permissions not directly justified by the skill's documented operations.

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Static analysis

No suspicious patterns detected.