ClawHub
Back to skill

Security audit

Gbrain Multi-Agent Search

Security checks for vulnerabilities and agentic risk

Overview

This skill is purpose-built for a private knowledge base, but it can access sensitive personal content and encourages remote embedding without a clear privacy warning.

Install only if you want an agent to search and manage your local gbrain/brain repositories, including personal notes, emails, documents, and diary entries. Use a local or self-hosted embedding backend for private material when possible, avoid storing a real API key in persistent instruction files, and require explicit confirmation before running sync, embed, link, or other mutation commands.

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 (4)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs users to configure an OpenAI-compatible embedding API and later run embedding commands, but it does not warn that embeddings may transmit personal knowledge-base content to an external service. In this context, the indexed data explicitly includes emails, diary entries, and personal documents, so omission of a clear disclosure creates a significant privacy and data-governance risk.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger list contains broad phrases such as "gbrain," "personal knowledge," and "knowledge base search" that can match ordinary user requests and cause the skill to activate unexpectedly. Because this skill can access highly sensitive local notes, emails, diary entries, and documents, over-broad activation increases the chance of unintentional data access or disclosure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
83% confidence
Finding

Referencing an external OpenAI-compatible API endpoint indicates that the skill is designed to send data outside the local environment during embedding operations. In a skill centered on personal notes and emails, that external transmission is security-relevant because users may assume the workflow is fully local unless clearly warned otherwise.

Content

Scanner excerpt · SKILL.md (reported line 25)May include surrounding context.

md
| `GBRAIN_DIR` | Path to gbrain source repo | `~/gbrain` |
| `BRAIN_DIR` | Path to brain content repo | `~/brain` |
| `OPENAI_API_KEY` | Embedding API key | `sk-...` |
| `OPENAI_BASE_URL` | Embedding API base URL | `https://api.openai.com/v1` |

### One-liner setup (add to TOOLS.md)

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The one-liner setup normalizes use of a remote API base URL and API key without any nearby privacy warning, making it easy for users to configure external transmission by copy-paste. Given the sensitive nature of the referenced knowledge base, this increases the likelihood of inadvertent disclosure to a third-party service.

Content

Scanner excerpt · SKILL.md (reported line 33)May include surrounding context.

export GBRAIN_DIR=/gbrain export BRAIN_DIR=/brain export OPENAI_API_KEY=your-key-here export OPENAI_BASE_URL=https://api.openai.com/v1 alias gb="cd $GBRAIN_DIR && bun run src/cli.ts"

text

Static analysis

No suspicious patterns detected.