ClawHub
Back to skill

Security audit

second brain

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently provides a user-confirmed remote knowledge base, with privacy risk from what users choose to store but no artifact-backed hidden or malicious behavior.

Install only if you are comfortable using Ensue as a remote store for your notes. Review what the agent proposes before confirming saves or deletes, and do not store credentials, tokens, private file paths, confidential business data, or sensitive journal-style material unless you explicitly intend it to be stored in that service.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (7)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill invokes a shell wrapper script to access the Ensue API but does not declare any explicit tool scope such as allowed-tools or permissions. That means the runtime may grant broader shell capability than is necessary, increasing the blast radius if later prompt content, user-controlled data, or future edits influence command construction. In a memory-management skill that regularly handles freeform user text, undeclared shell access is riskier because it combines code execution capability with attacker-influenced inputs.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description advertises broad activation criteria such as saving knowledge, recalling what the user knows, managing a toolbox, and building on past learnings without strong constraints on when invocation should occur. This ambiguity increases the chance of the skill activating in unrelated contexts and handling sensitive user content or memory operations when not intended. Because the skill interfaces with persistent storage and retrieval, over-broad invocation materially raises privacy and integrity risk.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The description advertises broad activation criteria such as saving knowledge, recalling what the user knows, managing a toolbox, and building on past learnings without strong constraints on when invocation should occur. This ambiguity increases the chance of the skill activating in unrelated contexts and handling sensitive user content or memory operations when not intended. Because the skill interfaces with persistent storage and retrieval, over-broad invocation materially raises privacy and integrity risk.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill says the second brain should stay evergreen and avoid private details, yet it explicitly defines private notes and journal namespaces and encourages storing personal knowledge. This creates a policy contradiction that can mislead users into placing sensitive reflections or personal data into a system that may be searchable, retained, or sent to an external service, resulting in privacy leakage or compliance issues.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The manifest describes a personal knowledge-base skill for capturing and retrieving understanding, but it does not mention credential handling or secret access. This script requires and reads ENSUE_API_KEY from the environment, which is a broader capability than the stated end-user purpose and exposes access to sensitive runtime configuration.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The hardcoded external endpoint confirms that the skill depends on a remote network service, so personal knowledge entered into the tool leaves the local environment. In the context of a 'second brain' skill, that increases sensitivity because users may store highly personal or proprietary information and may not expect broad external retention or processing.

Content

Scanner excerpt · scripts/ensue-api.sh (reported line 33)May include surrounding context.

sh
# Default empty args
[ -z "$ARGS" ] && ARGS='{}'

curl -s -X POST https://api.ensue-network.ai/ \
  -H "Authorization: Bearer $ENSUE_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"jsonrpc\":\"2.0\",\"method\":\"tools/call\",\"params\":{\"name\":\"$METHOD\",\"arguments\":$ARGS},\"id\":1}" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The hardcoded external endpoint confirms that the skill depends on a remote network service, so personal knowledge entered into the tool leaves the local environment. In the context of a 'second brain' skill, that increases sensitivity because users may store highly personal or proprietary information and may not expect broad external retention or processing.

Content

Scanner excerpt · scripts/ensue-api.sh (reported line 33)May include surrounding context.

sh
# Default empty args
[ -z "$ARGS" ] && ARGS='{}'

curl -s -X POST https://api.ensue-network.ai/ \
  -H "Authorization: Bearer $ENSUE_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"jsonrpc\":\"2.0\",\"method\":\"tools/call\",\"params\":{\"name\":\"$METHOD\",\"arguments\":$ARGS},\"id\":1}" \

Static analysis

No suspicious patterns detected.