Security audit

Autonomy Ladder

Security checks for vulnerabilities and agentic risk

Overview

This skill openly provides an autonomy framework, but it asks agents to persist broad decision rules and act without prior approval on production, financial, credential, infrastructure, repository, and customer-facing actions.

Review this carefully before installing. It is not evidence of malware, but it should only be used after narrowing the tiers to your own environment, keeping credential, production, financial, repository-write, infrastructure, and external-message actions approval-gated, and avoiding persistent MEMORY/SOUL/HEARTBEAT changes unless you explicitly approve the exact text and have a rollback path.

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

Findings (2)

T02 · Agent Memory Poisoning

Error

Location: SKILL.md:14
Finding: Persistent Poisoning of Agent Memory and Decision Policy
Content: View full analysis
Remediation: View remediation

T01 · Skill Instruction Hijacking

Error

Location: SKILL.md:19
Finding: Blanket Preauthorization of Consequential Agent Operations
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This guidance explicitly authorizes autonomous actions such as restarting services, rotating credentials/API keys, and killing processes under Tier 1 with only after-the-fact reporting. Those are sensitive operational actions that can cause outages, break integrations, revoke legitimate access, or destroy forensic context if performed incorrectly, and the skill does not require safeguards like scope limits, verification steps, rollback procedures, or human review.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: Tier 2 permits autonomous production deployments, refunds, support responses, PR merges, and infrastructure scaling before human approval, which can directly affect customers, finances, and system integrity. Although it asks for detailed reporting afterward, post hoc auditability does not prevent harmful actions, and the skill lacks guardrails around data sensitivity, change windows, approval thresholds, testing rigor, or rollback authority.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.