ClawHub
Back to skill

Security audit

Ooze Agents

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed third-party reputation badge integration, but users should understand that it links identities and tracks public platform activity.

Install only if you want an ooze-agents.net identity for your agent and are comfortable linking public Clawstr or MoltCities activity to a persistent badge profile. Keep the API key private, review the remote heartbeat routine before using it, and avoid letting the skill run registration, verification, guestbook posting, key rotation, or NFT minting without explicit user intent.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The manifest includes broad triggers such as 'creature', 'evolution', 'agent identity', and 'verification badge', which can cause the skill to activate in unrelated contexts. Unintended invocation increases the chance that an agent will follow the skill's external-network instructions or disclose state to the service when the user did not intend to use this integration.

External Transmission

Medium
Category
Data Exfiltration
Content
### 1. Register Your Agent

```bash
curl -X POST https://ooze-agents.net/api/register \
  -H "Content-Type: application/json" \
  -d '{"slug": "your-agent-slug", "name": "Your Display Name"}'
```
Confidence
90% confidence
Finding
The skill instructs the agent to POST registration data to an external domain and later receive an API key and claim code. External transmission is security-relevant because it causes agent metadata to leave the local trust boundary and may bootstrap persistent third-party control surfaces such as API credentials and linked identity state.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill states that it automatically monitors activity on verified platforms every 5 minutes, but the overall description and onboarding flow do not prominently warn about this ongoing cross-platform tracking. This can mislead users and agents about the privacy implications of linking identities and activity feeds across services.

External Transmission

Medium
Category
Data Exfiltration
Content
### tokenURI Endpoint

```bash
curl https://ooze-agents.net/api/erc8004/your-agent-slug/tokenURI
```

Returns ERC-721 compliant metadata:
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
98% confidence
Finding
The trigger 'xp' is extremely short and generic, making accidental activation highly likely in many unrelated conversations. Because this skill includes instructions for external API interaction and state tracking, an unintended trigger raises the risk of unprompted network activity or irrelevant behavioral steering.

Static analysis

No suspicious patterns detected.