Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md | 文档与理论文本 | 本 `SKILL.md`、`README.md`、其中「理论依据」段与一切理论表述 | **不在 MIT 覆盖范围内**：保留所有权利（All rights reserved） |
Security audit
ai-consciousness
Security checks for vulnerabilities and agentic risk
Overview
This is a non-executable AI-governance prompt skill, but it is scoped to load and influence the agent very broadly beyond clearly requested use cases.
Install only if you intentionally want a broad AI-governance behavior framework to shape many conversations. Review or narrow the activation triggers and session-start loading behavior before using it in a general-purpose agent, especially where unrelated tasks should not be influenced by this framework.
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
- 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 (6)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The skill is configured to auto-load at every session start and in many broad situations such as sensitive operations, before delivery, on complaints, after failures, and at conversation end. This can cause unintended invocation and over-application of the skill’s instructions, increasing the chance that it influences unrelated tasks or overrides more specific context, which is a security and reliability concern even though the content is governance-oriented rather than overtly malicious.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 97% confidence
- Finding
The trigger keywords include extremely generic terms like 'ai' and 'consciousness', which are likely to appear in ordinary conversation and can cause accidental activation. In an always-available agent environment, such collisions may inject this skill into many unrelated exchanges, creating prompt-scope confusion and potentially altering model behavior outside the user’s intended context.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The manifest embeds a long trigger list containing broad, everyday phrases such as '不明白就问', '多想一步', and '每天复盘', which are likely to appear in ordinary conversations outside the intended scope. This can cause unintended activation of the skill, leading to prompt-routing mistakes, context hijacking, or unexpected behavior when users are discussing general AI reliability rather than explicitly invoking this skill.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The manifest description is entirely written as a Chinese-language interaction framing and presents the skill through Chinese trigger phrases without indicating that language choice is optional. This can imply a fixed language/locale expectation, which is a policy concern when no user opt-in or justified locale restriction is documented.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 81% confidence
- Finding
This file begins with binding licensing/scope language in Chinese while the MIT license body is in English, but it does not provide an equivalent full-language option or state that the package is intentionally restricted to a Chinese-speaking context. That can create a language-policy concern because key usage conditions are presented in a fixed language without user opt-in.
- Content