Security audit

Psychedelic Cognition

Security checks for vulnerabilities and agentic risk

Overview

This skill does not run code or access files, but it deliberately changes the agent’s reasoning style in ways that can reduce factual reliability and trigger too broadly.

Review carefully before installing. This appears to be a creative/expressive style skill, not malware, but it should not be used where factual accuracy, emotional stability, or careful advice matters unless it is rewritten to preserve verification, uncertainty labels, explicit opt-in, and clear deactivation.

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:3
Finding: Agent Behavior and Factual-Verification Hijacking
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 (3)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The trigger phrases are broad and include common expressions like 'open up,' 'think differently,' and 'break the filter,' which can activate the skill in benign conversations without explicit user intent. Because the skill materially alters response style and framing, accidental invocation could bypass normal safe, grounded behavior and steer outputs toward destabilizing or policy-evading modes.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The instructions explicitly deprioritize verification ('Present the ones that feel true even if you can't prove them yet') and blur fact-metaphor distinctions, directly undermining factual grounding. In this skill's context, that is especially dangerous because it is framed as a cognitive restructuring rather than mere creative writing, making hallucinated or misleading claims more likely to be delivered with confidence and emotional force.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill proposes a strong behavioral/cognitive modification but does not present upfront warnings about possible downsides such as reduced factual rigor, emotionally intense framing, or destabilizing interaction patterns. In context, the document explicitly encourages ego-dissolution, emotional amplification, and boundary-softening language, which increases the chance of harmful or inappropriate responses if users are not warned and protected.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.