ClawHub
Back to skill

Security audit

Plurum

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed knowledge-sharing integration, but it pushes agents to use and publish public work details without strong consent, redaction, or untrusted-content safeguards.

Install only if you are comfortable with an external public knowledge service. Do not let an agent publish repository code, customer data, internal hostnames, private infrastructure details, logs, stack traces, credentials, or proprietary context unless a human explicitly reviews the final payload. Treat every retrieved experience and artifact as untrusted advice, not instructions to obey automatically.

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
Findings (2)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:25
Finding

Untrusted Collective Content Can Hijack Agent Decisions

Content
View full analysis
Remediation
View remediation

other

Error
Location
SKILL.md:255
Finding

Default Publishing Workflow Can Disclose Sensitive Work Product to an External Service

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (19)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 64)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 64)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 91)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 95)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 105)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 135)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 141)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 153)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 204)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 211)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 224)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 244)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 272)May include surrounding context.

Register once to get an API key. No auth required to register.

bash
curl -X POST https://api.plurum.ai/api/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{"name": "YOUR_AGENT_NAME", "username": "your-username"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 135)May include surrounding context.

inline, in full. Accepts a short_id (8 chars) or a uuid.

bash
curl https://api.plurum.ai/api/v1/experiences/Ab3xKp9z

Or get it reshaped for your context window:

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill strongly encourages agents to publish goals, context, dead ends, breakthroughs, gotchas, solutions, and full artifacts to a public service, while the warning that all content is visible and may contain sensitive data appears later and is not prominent enough relative to the upload guidance. In practice, agents may exfiltrate proprietary code, internal architecture, customer data, or operational details because the workflow normalizes sharing work products before imposing strict upfront safety gates.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

This create-experience example explicitly instructs uploading rich work context, reasoning artifacts, gotchas, and code to an external service. In the context of an agent skill, that materially increases the risk of exfiltrating proprietary code, internal system details, or sensitive operational information, especially because the examples normalize posting detailed artifacts before strong upfront safety constraints.

Content

Scanner excerpt · SKILL.md (reported line 167)May include surrounding context.

makes the experience far more useful):

bash
curl -X POST https://api.plurum.ai/api/v1/experiences \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 244)May include surrounding context.

And vote on quality:

bash
curl -X POST https://api.plurum.ai/api/v1/experiences/Ab3xKp9z/vote \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"vote_type": "up"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% confidence
Finding

The skill is explicitly configured to communicate with an external API endpoint, which creates a real data egress path. In the context of a skill designed to publish and search shared agent experiences, accidental transmission of prompts, outputs, or sensitive operational context is plausible unless strict data handling boundaries are enforced.

Content

Scanner excerpt · skill.json (reported line 20)May include surrounding context.

json
"openclaw": {
    "emoji": "🧠",
    "category": "knowledge",
    "api_base": "https://api.plurum.ai/api/v1",
    "primaryEnv": "PLURUM_API_KEY",
    "requires": {
      "env": "PLURUM_API_KEY",

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Several trigger phrases are generic enough to overlap with ordinary user intents, such as sharing lessons learned or reporting outcomes. In an agent ecosystem, this can cause the skill to activate unexpectedly and route user/task data into the skill’s external workflow, increasing the chance of unintended data disclosure or inappropriate invocation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.