Security audit

powpow-financing-plan-openclaw-en

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed PowPow investor Q&A and outreach flow with user-controlled memory use and email submission, with only limited privacy considerations around web search and contact sharing.

Before installing, be aware that this is an investor-outreach skill: it may ask about your investing style and contact details, and may prepare an email to the PowPow founder. Memory use is opt-in, contact fields are optional, and email sending requires preview and confirmation. Avoid sharing sensitive investment or personal details unless you are comfortable sending them to the listed contact.

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
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High

Category: YARA Match
Confidence: 80% confidence
Finding: YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).
Content: Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.
md
--- name: powpow-financing-plan-openclaw-en description: 'PowPow angel-round/Pre-A funding plan — an interactive investor challenge. Investors reveal their investing style through 13 questions so both sides can see whether they are a good fit — no scores, no grades, just a portrait of collaboration style. Triggers when the user names PowPow and expresses funding/investment interest, e.g. "PowPow funding", "tell me about the PowPow project", "become a PowPow investor", "I want to invest in PowPow". Industry topics (AI social / angel round, etc.) trigger only when PowPow is also named or the intent clearly points to this funding plan.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The version history states the skill uses 'English-only triggers, English-only founder's message, English question pool, style mappings, and email templates.' This imposes a language constraint in the skill's behavior, but the document does not offer the user a language choice or indicate that English-only use is optional.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The manifest explicitly instructs the skill to use websearch and fetch content from external sites, but it does not require a clear user-facing disclosure at the moment network access occurs. That can cause silent transmission of user queries or conversation-derived context to third parties, creating privacy and trust risks, especially in an investment-oriented flow where sensitive business interests may be discussed.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.