Back to skill

Security audit

Work Productivity Self Improving Workflow Helper

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with high confidence: it found no security issues in the completed checks, while noting broad activation wording that could invoke it for unrelated requests.

The verdict is benign, and the scan covered all six listed components. Consider whether broad automatic activation fits your preferences; the scanner noted that general terms could trigger the skill outside its intended workflow context.

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)

Generic activation keywords can trigger the skill on unrelated everyday requests.

Medium
Category
Trigger Abuse
Confidence
99% confidence
Finding

The trigger list includes broad words such as “self,” “learning,” “criticism,” and “organizing,” which commonly occur outside self-improving agent workflows. The description on line 4 also treats generic terms as activation conditions, and no exclusions or mandatory agent-workflow context constrain these matches.

Content

No source excerpt is available for this finding.

Generic trigger keywords can activate the skill for unrelated requests.

Medium
Category
Trigger Abuse
Confidence
99% confidence
Finding

The trigger list includes everyday terms such as self, learning, criticism, and organizing without requiring an agent-workflow context. Line 4 likewise treats several generic terms as activation conditions, and no exclusion conditions prevent unrelated requests from invoking the skill.

Content

No source excerpt is available for this finding.

Implicit invocation is enabled without a clear, task-specific activation scope.

Medium
Category
Trigger Abuse
Confidence
94% confidence
Finding

The description and default prompt discuss demand for self-improving and proactive workflows rather than identifying concrete user requests this skill handles. With implicit invocation enabled, there are no clear activation conditions or exclusions to distinguish this skill from general productivity assistance.

Content

No source excerpt is available for this finding.

Usage keywords are overly broad and lack activation boundaries.

Medium
Category
Trigger Abuse
Confidence
94% confidence
Finding

The usage signals include everyday terms such as "self," "learning," "organizing," and "bug fix," which overlap with many unrelated requests. No exclusion conditions or requirement to combine these terms with the specific self-improving agent context limits unintended invocation.

Content

No source excerpt is available for this finding.

Activation keywords include common words without contextual restrictions.

Low
Category
Trigger Abuse
Confidence
95% confidence
Finding

The Usage section lists broad keywords such as self, learning, criticism, and organizing that occur in everyday requests unrelated to this skill. Although specific trigger sentences follow, the file does not clarify that keywords alone are insufficient for activation or provide exclusion conditions.

Content

No source excerpt is available for this finding.

Activation keywords include broad everyday terms without scope restrictions.

Low
Category
Trigger Abuse
Confidence
95% confidence
Finding

The usage section lists keywords such as self, learning, organizing, and bug fix, which occur in many requests unrelated to this skill's self-improving agent workflows. No exclusion conditions or contextual requirements limit these keywords, creating a risk of unintended activation.

Content

No source excerpt is available for this finding.