ClawHub
Back to skill

Security audit

cve-sentinel

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed CVE-monitoring helper that queries official vulnerability databases and writes local watchlist/report files, with no evidence of hidden persistence or unrelated data exfiltration.

Install only if you are comfortable with the skill reading dependency manifest files in the project directory you choose, sending dependency/CVE lookup data to official vulnerability services, and writing watchlist/report files locally. Prefer a restricted project path, review output paths before running, and avoid passing an NVD API key on the command line; use an environment or secret-management approach if you adapt the script.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/query-vulns.py:55
Finding

NVD API Key Exposed Through Command-Line Arguments

Content
View full analysis
list: start_utc = start_date.astimezone(timezone.utc) if start_date.tzinfo else start_date.replace(tzinfo=timezone.utc) end_utc = end_date.astimezone(timezone.utc) if end_date.tzinfo else end_date.replace(tzinfo=timezone.utc) params = { "keywordSearch": keyword, "resultsPerPage": 100, } headers = {} if api_key: headers["apiKey"] = api_key delay = NVD_DELAY_WITH_KEY if api_key else NVD_DELAY_NO_KEY ``` The same behavior occurs for CVE-specific queries: ```python def query_nvd_cve(cve_id: str, api_key: str = Non ...[truncated 2551 chars]
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The skill is presented as CVE monitoring and analysis, but it also performs local dependency extraction from a project directory and modifies watchlist.yaml. This mismatch can cause users to authorize what appears to be passive analysis while the skill actually inventories local project contents and changes files, which expands privacy and integrity risk beyond the stated purpose.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill instructs the agent to read and write local files and query external vulnerability sources, but it does not declare any explicit tool scope or allowed-tools boundary. That creates an overbroad capability surface where a caller may invoke filesystem and network actions without transparent permission constraints, increasing the risk of unintended data access, overwrite, or exfiltration during use.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/query-vulns.py (reported line 81)May include surrounding context.

python
url = f"{OSV_BASE}/v1/query"
    payload = {"package": {"name": name, "ecosystem": ecosystem}, "version": version}
    try:
        resp = requests.post(url, json=payload, timeout=90)
        resp.raise_for_status()
        data = resp.json()
        vulns = data.get("vulns", [])

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/query-vulns.py (reported line 312)May include surrounding context.

python
def query_github_advisory(cve_id: str) -> dict:
    url = "https://api.github.com/advisories"
    params = {"cve_id": cve_id}
    resp = _retry_request(requests.get, url, params=params, timeout=30)
    if resp is None:

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

This code creates parent directories and writes results to an arbitrary file path provided via --output, which is a file-modifying operation. While the argument help mentions an output file path, there is no explicit user-facing warning at the write site or comment/docstring highlighting that the script will create directories and overwrite the target file.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.