Work Productivity Tavily Web Workflow Helper

• Outcome: Pass
• Latest audit: Jul 24, 2026 · 5:08 PM UTC
• Version: 0.1.0

Overview

This is a static workflow-help skill with overbroad activation wording, but it does not contain executable code, credential handling, persistence, or hidden data movement.

Before installing, consider narrowing or disabling implicit activation so ordinary web/search/API requests do not accidentally invoke this skill. The inspected package appears safe as a workflow guide, but its triggers are broader than ideal.

SkillSpector

By NVIDIA

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 (10)

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The trigger sentence is broad and natural-language-like enough that ordinary user requests could unintentionally activate this skill. In an agent environment, overbroad activation can route unrelated tasks into a web-search workflow, causing unintended external queries, data exposure to third-party services, or unsafe tool use without clear user intent.

Vague Triggers

Medium

• Confidence: 82% confidence
• Finding: Although additional context is present, the trigger still remains ambiguous because it describes a generic need for a 'practical workflow' rather than a clearly bounded skill invocation. This increases the chance of misfires, where the agent may apply this skill in contexts involving sensitive troubleshooting or setup details and unnecessarily involve web/search actions.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The trigger phrases are broad enough to match ordinary user requests about workflows, bug fixing, hardening, or practical help, which can cause this skill to activate when the user did not specifically intend to invoke it. In an agent ecosystem, over-broad activation can route unrelated requests into a web-search-oriented workflow, increasing the chance of unintended tool use, prompt interference, or confusing outputs.

Vague Triggers

High

• Confidence: 93% confidence
• Finding: The manifest description is extremely broad and includes generic terms like 'work-productivity', 'web', and 'practical workflow', which can cause the skill to be selected for many unrelated user requests. Over-broad activation increases the chance that this skill intercepts prompts outside its intended scope, leading to incorrect delegation, prompt-surface expansion, and possible propagation of unsafe or irrelevant instructions.

Vague Triggers

High

• Confidence: 95% confidence
• Finding: The keyword trigger list contains vague, high-frequency terms such as 'web', 'via', 'user', and 'api', which are likely to match a wide range of unrelated conversations. This makes unintended invocation much more likely and can route sensitive or irrelevant tasks through this skill, increasing attack surface and reducing reliability of skill selection.

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The example trigger phrases are phrased as ordinary conversational requests and can match common speech patterns without clear boundaries. Ambiguous examples train or bias invocation systems toward over-selection, which can activate the skill in contexts where it is not appropriate and create confusion or unsafe task routing.

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The trigger list is overly broad and includes generic terms like 'web', 'search', 'api', 'user', and 'bug fix', which are common in unrelated conversations. This can cause accidental invocation of the skill in contexts it was not intended for, leading to confusing behavior, misrouting, or unintended access to workflow logic that may influence user actions without clear relevance.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The default prompt is extremely broad and overlaps with ordinary user requests about productivity, Tavily, web, search, workflows, checklists, analysis, and implementation support. This creates a real risk of unintended or overbroad activation, causing the skill to inject its prompt and behavior into conversations where the user did not clearly request this specific skill.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: Enabling implicit invocation without narrow activation criteria allows the skill to be auto-selected based on vague semantic overlap rather than explicit user intent. In a broadly described productivity/search skill, this increases the chance of prompt injection into unrelated tasks, response steering, and unintended tool or workflow influence.

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The trigger sentences and keywords are broad enough to match common user requests about work, web search, APIs, bug fixes, or practical workflows, which can cause the skill to activate outside its intended scope. Overbroad activation increases the chance of prompt/skill hijacking at the routing layer, misapplication of the skill to unrelated tasks, and reduced user control over which capability is invoked.

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.