Security audit

Adaptive Skill Factory

Security checks for vulnerabilities and agentic risk

Overview

The skill is not malicious, but it can implicitly install, create, update, or delete agent skills with broad triggers and limited user-control guidance.

Install only if you want the agent to actively manage skills for recurring workflows. Before use, require explicit confirmation for installing, creating, updating, or deleting skills, and avoid shared paths like D:\internal-hub\skills unless you intend cross-role persistent changes.

A.I.G

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

SkillSpector

Vulnerability Patterns

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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Ae1

High

Category: analysis-evasion
Content: - Keep `SKILL.md` concise and procedural.
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The description says to use the skill when "a workflow gap appears" and includes broad conditions like "a task is repetitive" or when "an existing skill might save time." These activation cues are open-ended and lack explicit constraints or negative examples, which could cause the skill to be invoked in many ordinary situations rather than a narrowly defined context.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The default prompt uses very broad routing language ('decide whether this task should reuse, install, or create a skill'), which can match many ordinary tasks and cause the skill to be invoked outside narrow, intended conditions. In combination with allow_implicit_invocation: true, this increases the chance of unintended autonomous skill discovery/installation/creation, expanding the attack surface and enabling risky prompt steering through normal workflow language.

Static analysis

No suspicious patterns detected.