ClawHub
Back to skill

Security audit

Blindoracle Fixed

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed API-integration guide for sending agent audit and passport data to a user-configured BlindOracle/MASSAT endpoint.

Install only if you trust the MASSAT_API_URL service you configure. Use HTTPS, a dedicated least-privilege BLINDORACLE_API_KEY, and avoid sending unnecessary agent or operator details.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 76)May include surrounding context.

Run a security audit against an agent

bash
curl -X POST "$MASSAT_API_URL/api/v1/audit" \
  -H "Authorization: Bearer $BLINDORACLE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

The skill instructs users to send agent metadata and an API bearer token to a user-configured external endpoint via curl. This is a real data egress path: if the endpoint is misconfigured, malicious, or compromised, sensitive operational metadata and credentials can be exposed or abused. The risk is moderated by the fact that the documentation explicitly discloses the transmission and frames it as core functionality rather than hidden exfiltration.

Content

Scanner excerpt · SKILL.md (reported line 76)May include surrounding context.

Run a security audit against an agent

bash
curl -X POST "$MASSAT_API_URL/api/v1/audit" \
  -H "Authorization: Bearer $BLINDORACLE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The passport issuance example transmits identity-related agent data plus a bearer API key to an external service. Because this operation handles identity/passport issuance, compromise of the remote endpoint or token could enable unauthorized registrations, identity abuse, or leakage of operator information. The context makes this expected behavior for the skill, but it remains a genuine security-sensitive external transmission.

Content

Scanner excerpt · SKILL.md (reported line 96)May include surrounding context.

Register an agent with ERC-8004 passport

bash
curl -X POST "$MASSAT_API_URL/api/v1/passport/issue" \
  -H "Authorization: Bearer $BLINDORACLE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{