ClawHub
Back to skill

Security audit

Exa

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Exa API helper that sends user-provided search queries or URLs to Exa, with no hidden persistence or unrelated local access found.

Install this only if you are comfortable sending search queries, code-search prompts, and requested URLs to Exa using your EXA_API_KEY. Avoid putting secrets, private source snippets, internal URLs, or confidential project details into queries unless your organization permits that use of Exa.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill instructs users to execute local shell scripts (bash scripts/search.sh, code.sh, content.sh) but does not declare any permissions indicating shell/code execution capability. This creates a transparency and policy-enforcement gap: consumers or orchestrators may treat the skill as lower risk than it really is, while it can invoke arbitrary local commands through bundled scripts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script sends the user-supplied query to Exa's external API, but the script itself does not disclose at execution time that input content will leave the local environment. In an agent setting, users may pass sensitive code snippets, internal identifiers, or proprietary questions, creating an unintended data exfiltration/privacy risk even though the network behavior is part of the skill's purpose.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
88% confidence
Finding

The hardcoded Exa API endpoint confirms that this skill sends data off-host to an external provider. In the context of an agent skill that may be used on code and documentation, this increases the chance that proprietary source, internal URLs, or sensitive prompts are transmitted outside the trust boundary.

Content

Scanner excerpt · scripts/code.sh (reported line 31)May include surrounding context.

sh
highlights: true
    }')

curl -s -X POST 'https://api.exa.ai/search' \
    -H "x-api-key: $EXA_API_KEY" \
    -H 'Content-Type: application/json' \
    -d "$PAYLOAD"

External Transmission

Medium
Category
Data Exfiltration
Confidence
88% confidence
Finding

The hardcoded Exa API endpoint confirms that this skill sends data off-host to an external provider. In the context of an agent skill that may be used on code and documentation, this increases the chance that proprietary source, internal URLs, or sensitive prompts are transmitted outside the trust boundary.

Content

Scanner excerpt · scripts/code.sh (reported line 31)May include surrounding context.

sh
highlights: true
    }')

curl -s -X POST 'https://api.exa.ai/search' \
    -H "x-api-key: $EXA_API_KEY" \
    -H 'Content-Type: application/json' \
    -d "$PAYLOAD"

Static analysis

No suspicious patterns detected.