Back to skill

Security audit

golang-gopls

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Go language-server helper for navigation, refactoring, diagnostics, and limited vulnerability checks, with no evidence of hidden or malicious behavior.

Install this only if you want an agent to use gopls on your Go projects. Expect it to inspect local Go code and dependencies, run diagnostics/tests, possibly contact Go module services during build resolution, and modify source files when you ask for refactors or formatting. Review diffs before accepting CLI write flags or raw execute/code-action operations.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 49)May include surrounding context.

md
ntation, or trace a call graph before touching code you didn't write. Details: [references/features.md](references/features.md#navigation).

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 54)May include surrounding context.

md
ntation, or trace a call graph before touching code you didn't write. Details: [references/features.md](references/features.md#navigation).

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The manifest frames this skill as semantic code intelligence and explicitly says whole-tree vulnerability audit should use a different skill. However, the body documentation directs the agent to run go_vulncheck once at session start and again after go.mod changes, expanding behavior into vulnerability scanning beyond the primary navigation/refactoring scope.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This is an internal intent contradiction in the skill documentation: the top-level description excludes vulnerability-audit use, but later sections instruct unconditional baseline go_vulncheck and whole-workspace scans after dependency changes. That is more than an omission; it actively sends mixed signals about whether vulnerability analysis is in-scope.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

This markdown file documents flags and commands that can overwrite source files (-w) and execute code actions or commands, but it does not include a clear caution statement about the risk of unintended file changes before presenting usage. While the behavior is described functionally, the skill description lacks an explicit warning about impact to user data/system state, which is required for markdown files when behaviors can affect files.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The execute command is described as sending a raw LSP ExecuteCommand request and examples include test-running and import-modifying actions, but there is no explicit safety warning about side effects or workspace changes. For markdown guidance, operations that may affect user data or system integrity should be accompanied by a clear caution.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The manifest says this skill is 'Not for a whole-tree vulnerability audit' and directs users to a separate golang-security skill using govulncheck. However, this file documents a go_vulncheck MCP tool and recommends running it on ./..., which is a vulnerability-checking capability that conflicts with the stated scope boundary.

Content

No source excerpt is available for this finding.