ClawHub
Back to skill

Security audit

Partnership Revenue Engine

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed business playbook for partnership and channel revenue work, with no hidden code, persistence, or destructive behavior found.

Reasonable to install if you want a partner-program playbook. Review generated outreach, agreements, newsletters, and QBR materials before using them externally, and be mindful that partner research may rely on public web data that should be verified.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
rules:
    - No brand bidding on paid search
    - No coupon/deal sites without approval
    - Honest representation required
    - FTC disclosure mandatory
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.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The command promises to build a partner research brief 'from web data' without clearly disclosing that the skill may browse external sources, collect third-party information, or rely on untrusted internet content. In an agent setting, this can trigger unexpected network access and data handling beyond what a user may reasonably expect, increasing privacy, compliance, and prompt-injection exposure.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger 'Build enablement kit for [partner name]' is broad enough to overlap with ordinary business conversation, making accidental invocation plausible. In agent environments, ambiguous triggers can cause unintended execution of a multi-step workflow that may generate materials, process stored partner data, or take actions without clear user intent.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The phrase 'Design partner program tiers' is highly generic and can easily appear in normal planning discussions, which raises the risk of unintended skill activation. If auto-invoked, the skill could produce policy-like outputs or overwrite user expectations without a deliberate request to run the skill.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger 'Draft partner newsletter for this month' is ambiguous and resembles a routine instruction a user might say outside the context of invoking this specific skill. That ambiguity increases the likelihood of accidental execution and unintended use of partner data, prior materials, or automated content generation.

Scope Creep

Low
Category
Excessive Agency
Content
**Score = weighted sum × 4 = max 100**

- **80-100:** Champion — invest more, expand scope
- **60-79:** Healthy — maintain cadence, look for growth
- **40-59:** At risk — diagnose issues, intervention plan
- **<40:** Declining — honest conversation, consider winding down
Confidence
75% confidence
Finding
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Scope Creep

Low
Category
Excessive Agency
Content
| Onboarding | Enablement + training | Completion of 14-day checklist | 2 weeks |
| Ramping | First referrals + support | First closed deal | 30-90 days |
| Producing | Consistent referral flow | Meets monthly targets | Ongoing |
| Scaling | Expand scope, co-sell | Revenue growing QoQ | 6+ months |
| Strategic | JVs, co-build, exclusivity | Board-level relationship | 12+ months |

---
Confidence
75% confidence
Finding
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Static analysis

No suspicious patterns detected.