Back to skill

Security audit

Spec Plan Build Review

Security checks for vulnerabilities and agentic risk

Overview

This skill is a transparent workflow guide for planning, building, verifying, reviewing, and shipping software work, with no hidden code or persistence.

Install this if you want the agent to apply a structured delivery loop to coding or skill work. Be aware it may be invoked for many ordinary software tasks, and for real releases you should still confirm commits, pushes, publication, and CI status before treating work as shipped.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: spec-plan-build-review
description: "Run a proportional delivery lifecycle for software or skill work: clarify scope, create a concise plan, implement, verify, review, and ship. Use when the user asks to spec, plan, build, test, review, ship, release, prepare a PR, prepare ClawHub or GitHub publication, or coordinate a multi-step coding task that should not skip verification."
---

# Spec Plan Build Review

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The invocation description says to use the skill when the user asks to "plan," "build," "test," "review," or "ship," which are broad terms that commonly appear in ordinary software conversations. Without clearer boundaries, exclusions, or examples of when not to invoke the skill, this could cause unintended activation for many routine requests.

Content

No source excerpt is available for this finding.