ClawHub
Back to skill

Security audit

self-improving-business

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed business-learning logger with optional local reminder hooks and no evidence of hidden execution, exfiltration, or unauthorized business actions.

Install only if you are comfortable with an agent keeping persistent local business notes. Treat `.learnings/` as potentially sensitive, keep it out of shared repositories unless sanitized, and enable hooks only in the project where you want reminder behavior.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • 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
Findings (12)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The declared description presents an operational/business-analysis skill meant to capture and surface administration problems and trigger on specific organizational conditions. The supplied code does not analyze, capture, classify, or respond to such business issues. Its primary purpose is developer tooling: generating a scaffold for a new skill markdown file in a safe output directory. While the template text references business skills, the actual behavior is file creation and validation, which is materially different from the declared purpose. Therefore this is a clear description-behavior mismatch.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 517)May include surrounding context.

md
Extracted skills are untrusted until a human reviews the generated `SKILL.md`. Do not keep or publish an extracted skill without explicit user approval.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 539)May include surrounding context.

md
Extracted skills are untrusted until a human reviews the generated `SKILL.md`. Do not keep or publish an extracted skill without explicit user approval.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

Lines L012-L024 and L593-L637 characterize the skill as a passive logging/reminder mechanism with no execution behavior. However, the document later directs users to invoke executable scripts such as scripts/activator.sh, scripts/error-detector.sh, and especially scripts/extract-skill.sh (L479-L498, L510-L512, L533-L538), which goes beyond purely passive documentation and creates a contradiction in the stated intent.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
86% confidence
Finding

The skill directs creation of persistent files under ~/.openclaw/workspace/.learnings, which can outlive the current session and accumulate sensitive operational data across projects or contexts. Session persistence increases the risk of unintended retention, cross-project data mixing, and later disclosure through backups, syncing, or shared host access.

Content

Scanner excerpt · SKILL.md (reported line 84)May include surrounding context.

Manual clone is optional. Check out a reviewed tag or commit before use. Do not install an unpinned default branch into a shared environment.

Create Learning Files

bash
mkdir -p ~/.openclaw/workspace/.learnings

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 331)May include surrounding context.

md
| Priority | When to Use | Business Examples |
|----------|-------------|-------------------|
| `critical` | Material operational disruption, uncontrolled budget risk, or compliance exposure | Payroll-impacting process outage, major procurement exception without approval trail |
| `high` | SLA misses, delayed governance decisions, or significant KPI variance | Missed month-end close dependency, recurring approval bottleneck |
| `medium` | Important process hardening or definition alignment work | KPI glossary mismatch, outdated handoff checklist |
| `low` | Documentation cleanup and minor process optimization | Naming consistency updates, low-risk reporting cleanup |

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest describes a business-administration learning and issue-capture skill focused on improving operational decision quality. Automatically extracting and generating new reusable skills via extract-skill.sh is a separate meta-development capability that is not an obvious requirement for capturing policy gaps, KPI misalignment, or handoff failures.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
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.

Content

Scanner excerpt · SKILL.md (reported line 633)May include surrounding context.

md
### Ownership Rules
- This skill writes only to `.learnings/business/` in stackable mode.
- Do not read other skill folders, their SKILL.md files, or their log entries.
- Standalone mode writes to this project's `.learnings/*.md` log files only.
- Stackable mode writes only to the namespaced folder above and must not rewrite other skills' log entries.
- Promotion into `AGENTS.md`, `SOUL.md`, `TOOLS.md`, `MEMORY.md`, rules, hooks, or generated skills is not a logging write. Show a reviewed diff and apply only after explicit user approval.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · assets/LEARNINGS.md (reported line 32)May include surrounding context.

md
## Reminder-Only Boundary

Entries are documentation and reminders only.
No approvals, commitments, or transactions are executed from this file.

---

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · assets/SKILL-TEMPLATE.md (reported line 82)May include surrounding context.

md
## Reminder-Only Boundary

Entries are documentation and reminders only.
No approvals, commitments, or transactions are executed from this file.

---

Session Persistence

Medium
Category
Rogue Agent
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.

Content

Scanner excerpt · references/openclaw-integration.md (reported line 20)May include surrounding context.

Optional Hook Install

bash
mkdir -p .openclaw/hooks
cp -r hooks/openclaw .openclaw/hooks/self-improving-business

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill persistently writes operational learnings, governance issues, budget/procurement notes, and stakeholder details to local markdown files without an upfront sensitivity warning. In practice, this can cause users or agents to store confidential business context, audit issues, approval gaps, or vendor information in repositories or shared workspaces where it may later be synced, committed, or exposed.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.