ClawHub
Back to skill

Security audit

Focus Flow Optimizer Free

Security checks for vulnerabilities and agentic risk

Overview

This is a local personal productivity skill that writes simple planning files only with user confirmation, with some broad wording and team examples users should treat as scope notes.

Install only if you want a local Markdown-based productivity system under ~/focus-flow/. Review generated files before confirming writes, and keep team/member delegation content as personal planning notes rather than automated collaboration, HR, or performance-management authority.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The capability coverage statement lists an extremely broad set of keywords spanning general project management, planning, tracking, collaboration, and common productivity language. In an agent environment, this ambiguity can cause the skill to intercept many unrelated requests, leading to priority hijacking, unintended file operations, and application outside its documented trust and authorization model.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The activation scope includes generic expressions like feeling inefficient or overwhelmed, which overlap with ordinary conversation and many unrelated tasks. Overly broad triggers make accidental invocation more likely, so the agent may start applying this skill and writing or restructuring task data when the user did not intend to enter this workflow.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill is presented as a personal local productivity system, but its coverage text and examples extend into team-lead and delegation scenarios. That mismatch can cause the agent to apply the skill in broader contexts than intended, increasing the chance of inappropriate handling of coworker-related tasks, priorities, or performance-adjacent decisions without clear guardrails.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The documentation says the free version does not support team collaboration mode, yet earlier workflows instruct delegation to team members and use by a team负责人. This inconsistency can mislead an agent into performing unsupported collaboration behaviors anyway, causing overreach into multi-party workflows and creating unsafe assumptions about authorization and data handling.

Content

No source excerpt is available for this finding.