Work Productivity Ontology Typed Workflow Helper

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 10:36 AM UTC
• Version: 0.20260729.110214

Overview

This is a non-executable workflow helper with broad activation wording, but no evidence of malicious commands, data access, persistence, or deception.

Install only if you want a broad workflow helper for ontology or typed knowledge-graph style tasks. Be aware it may activate for generic terms like knowledge, graph, workflow, or bug fix, so review its use when a more specific skill would be better.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (14)

Vague Triggers

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill description says to use the skill whenever a user asks for broad concepts like work-productivity, knowledge, graph, workflow, checklist, analysis, or implementation support. Those terms are common across many unrelated tasks, so the skill can activate outside its intended scope and steer conversations unexpectedly. In an agent system, over-broad routing increases the chance of prompt-space hijacking, unintended delegation, and inappropriate application of this skill to sensitive or irrelevant requests.

Content

No source excerpt is available for this finding.

Vague Triggers

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

The keyword list includes ambiguous terms such as 'knowledge', 'graph', 'structured', 'memory', 'creating', and 'bug fix', which appear in a huge range of normal user requests. Generic keyword triggers make accidental activation likely, and attackers can easily invoke the skill by sprinkling common words into a prompt. Because this skill presents itself as a helper for workflows and implementation support, misrouting could influence downstream behavior in many contexts.

Content

No source excerpt is available for this finding.

Vague Triggers

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The trigger sentences are broad, natural-language phrases that can match ordinary user requests rather than a clearly scoped invocation condition. This increases the chance of unintended skill activation, which can route users into the wrong workflow, override more appropriate skills, or be abused to steer agent behavior through prompt-like phrasing.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The trigger phrases are broad, generic, and include common help-seeking language, which can cause the skill to activate for unrelated user requests. In an agent ecosystem, this creates routing confusion and raises the chance that users are unintentionally pushed into this workflow, leading to incorrect automation, misleading outputs, or accidental overshadowing of more appropriate skills.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The trigger phrases are generic and map to common requests like asking for practical workflows or help fixing bugs, which can cause the skill to activate outside its intended scope. In an agent ecosystem, overbroad activation can route unrelated user tasks into this skill, increasing the chance of inappropriate instructions, confusion, or unsafe automation being applied where it does not belong.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 94% confidence

Finding

The example trigger sentences use ordinary phrasing like 'Help me' and 'I need a practical workflow', which does not establish clear activation boundaries. This teaches routing systems or authors that everyday language is sufficient to call the skill, reinforcing over-triggering behavior already present in the description and keywords. While less severe than the broad keyword list itself, it still increases accidental invocation and makes abuse easier.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 93% confidence

Finding

The manifest-style description says to use the skill whenever a user asks about broad categories like work-productivity, ontology, typed, knowledge, or graph, or whenever they need practical workflow or analysis support. That description is overly expansive and ambiguous, making the skill eligible for many unrelated requests and increasing the chance of incorrect invocation or priority hijacking over narrower skills.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The trigger keyword list is very broad and includes generic terms such as "knowledge", "graph", "structured", and "creating", which can match many ordinary user requests unrelated to this skill’s specific purpose. This can cause unintended activation, routing users into the wrong workflow and potentially overriding more appropriate skills or causing confused-deputy style behavior in multi-skill environments.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The default prompt is framed as a broad, natural-language helper invocation and can match common user requests about productivity, workflows, checklists, or implementation support without clear boundaries. This increases the chance of accidental or overbroad skill activation, causing the agent to inject this skill into unrelated conversations and potentially override user intent or expand the attack surface for prompt-level abuse.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

Enabling implicit invocation without strict activation boundaries allows the platform to auto-select this skill based on vague semantic similarity rather than explicit user choice. In combination with the broad productivity/ontology framing, this can lead to unintended invocation in many benign contexts, increasing the risk of inappropriate prompt injection, mis-scoped assistance, or unintended data flow through the skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

• Category: Not specified by scanner
• Confidence: 72% confidence

Finding

The natural-language documentation explicitly defines only English and Chinese variants for instructions and user guides. This can be a locale/language policy concern if the skill behavior or documentation is implicitly limited to those languages without stating user choice or broader language support.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

• Category: Not specified by scanner
• Confidence: 78% confidence

Finding

This zh-CN README contains prominent English descriptive content for requirements and target users, but does not indicate that users can choose their preferred language. In a locale-specific file, forcing mixed-language instructions without opt-in can violate language/locale expectations.

Content

No source excerpt is available for this finding.

Vague Triggers

Low

• Category: Not specified by scanner
• Confidence: 89% confidence

Finding

The example trigger phrases are framed as generic help requests and include minimal domain-specific constraints, so they reinforce loose matching behavior instead of precise invocation. In systems that learn or rank from examples, such phrases can widen activation boundaries and increase accidental selection for unrelated tasks.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

• Category: Not specified by scanner
• Confidence: 75% confidence

Finding

The requirement rationale explicitly names source families including CSDN and SegmentFault, and the evidence list includes Chinese-language sources, but the skill description does not state whether users can choose their preferred language or locale. This can create a locale bias in how the skill frames evidence or outputs without explicit user opt-in.

Content

No source excerpt is available for this finding.