Security audit

导数取点

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed educational math skill for derivative root-existence proofs, with no evidence of hidden execution, data access, exfiltration, or unsafe persistence.

Review the language and curriculum fit before installing: the README and many examples are Chinese-focused, while the main skill also includes an English guide. The install commands are standard git-clone steps into skill folders; install only from a repository you trust.

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

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Rogue AgentSelf-Modification, Session Persistence
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (14)

Skill Enumeration

Medium

Category: Agent Snooping
Content: macOS / Linux： ```bash mkdir -p ~/.codex/skills git clone https://github.com/Jiuxiao-yunwai/derivative-point-picking.git ~/.codex/skills/derivative-point-picking ```
Confidence: 85% confidence
Finding: Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Session Persistence

Medium

Category: Rogue Agent
Content: macOS / Linux： ```bash mkdir -p ~/.codex/skills git clone https://github.com/Jiuxiao-yunwai/derivative-point-picking.git ~/.codex/skills/derivative-point-picking ```
Confidence: 60% confidence
Finding: Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Skill Enumeration

Medium

Category: Agent Snooping
Content: macOS / Linux： ```bash mkdir -p ~/.claude/skills git clone https://github.com/Jiuxiao-yunwai/derivative-point-picking.git ~/.claude/skills/derivative-point-picking ```
Confidence: 85% confidence
Finding: Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Session Persistence

Medium

Category: Rogue Agent
Content: macOS / Linux： ```bash mkdir -p ~/.claude/skills git clone https://github.com/Jiuxiao-yunwai/derivative-point-picking.git ~/.claude/skills/derivative-point-picking ```
Confidence: 60% confidence
Finding: Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Skill Enumeration

Medium

Category: Agent Snooping
Content: macOS / Linux： ```bash mkdir -p .claude/skills git clone https://github.com/Jiuxiao-yunwai/derivative-point-picking.git .claude/skills/derivative-point-picking ```
Confidence: 85% confidence
Finding: Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Natural-Language Policy Violations

Medium

Confidence: 91% confidence
Finding: The document title and structure present Chinese first and label the Chinese section as the main explanation, while no instruction states that language should be chosen based on user preference. Under the policy, forcing a specific language or locale without opt-in is a natural-language policy violation.

Natural-Language Policy Violations

Low

Confidence: 93% confidence
Finding: The skill README is entirely written in Chinese and explicitly labels the skill language as '中文'. Under the policy rule, forcing a specific language without offering the user a language or locale choice is a natural-language policy concern.

Vague Triggers

Low

Confidence: 88% confidence
Finding: This is a manifest file, so vague-trigger review applies. The default prompt describes using the skill for "this derivative problem" but does not define boundaries, exclusions, or a precise trigger scope, which can make invocation conditions ambiguous in broader educational contexts.

Natural-Language Policy Violations

Low

Confidence: 84% confidence
Finding: Natural-language policy review applies to all file types. The interface metadata is in Chinese, but the default prompt forces English phrasing ("Use $derivative-point-picking...") and does not offer the user any language or locale selection, which can be interpreted as a language-policy constraint without opt-in.

Natural-Language Policy Violations

Low

Confidence: 81% confidence
Finding: The document presents the core instructional content primarily in Chinese, with only the 'Key Pattern' section mirrored in English. This can be interpreted as enforcing a specific language experience without clearly offering a user language choice, which matches the policy category for language or locale constraints.

Natural-Language Policy Violations

Low

Confidence: 78% confidence
Finding: The file presents the problem and explanatory text in Chinese with an English subsection, but it does not indicate that language choice is optional or justified by a region-specific constraint. Under the language/locale policy rule, fixed language behavior without user opt-in can be a natural-language policy concern.

Natural-Language Policy Violations

Low

Confidence: 2% confidence
Finding: This markdown file contains both Chinese and English explanations, which does not force a specific language without user opt-in. The content is educational mathematics material and does not express any locale-restrictive policy.

Natural-Language Policy Violations

Low

Confidence: 84% confidence
Finding: The content explicitly presents the skill in Chinese and English, but there is no indication that language choice is user-selectable or that a bilingual requirement is justified by a region-specific constraint. Per the policy, imposing a language/locale behavior without opt-in can be a natural-language policy issue.

Natural-Language Policy Violations

Low

Confidence: 96% confidence
Finding: This markdown file is predominantly written in Chinese, with only a brief English title and closing sentence. Under the language/locale policy rule, forcing a specific language without user opt-in can be a natural-language policy violation when no justification or alternative is provided.

Static analysis

No suspicious patterns detected.