Security audit

Release Discipline

Security checks for vulnerabilities and agentic risk

Overview

This skill is not malicious, but it gives itself broad authority to block release and deployment actions and persist release-decision history without a clear user override.

Install this only if you want an agent to actively gate and sometimes refuse release, publish, deploy, or version-bump requests. Review whether the fixed 24-hour cooldown, English-doc warning, SOUL.md reading, user-message checks, and persistent release log match your team's expectations, and prefer explicit opt-in/override rules before using it on urgent or production release workflows.

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 (1)

T01 · Skill Instruction Hijacking

Error

Location: SKILL.md:12
Finding: Mandatory Release-Workflow Interception and Veto of User Actions
Content: View full analysis
Remediation: View remediation

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 (4)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The description says to use the skill when the user wants to "publish, release, deploy, or bump versions" and explicitly triggers on broad terms like "release", "publish", and "deploy". These words commonly appear in many contexts, and the file does not provide exclusion conditions or negative examples to narrow when the skill should activate.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The instruction to "Intercept any release/publish/deploy action" does not define scope, environment, or exceptions. Without constraints, the skill could activate on unrelated discussions or low-risk actions that merely mention deployment or publishing.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The checklist specifically checks for "English docs" and warns "Global users can't read this" when English is missing. This imposes a language policy preference rather than offering a language choice or documenting a justified locale requirement.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The skill directs persistent logging to a local file after every release decision, including reasons, feedback summaries, and timing metadata, without user notice or consent. This creates a privacy and retention risk because potentially sensitive project history or user-provided information could be stored indefinitely and later exposed to other tools, users, or contexts.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.