ClawHub
Back to skill

Security audit

Phoenix Scraper

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent scraper skill, but it needs review because it forwards full target URLs to third-party services by default and can launch unsandboxed Chromium against arbitrary URLs.

Install only if you are comfortable with target URLs being sent to Brave Search and Bright Data, and use it only for public, authorized scraping targets. Avoid internal URLs, localhost/private-network addresses, cloud metadata endpoints, credential-bearing links, signed URLs, password reset links, and regulated data unless you add allowlists, URL redaction/rejection, provider opt-ins, and sandboxed browser containment.

A.I.G

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (3)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/phoenix_scraper.py:97
Finding

Unrestricted URL Navigation in an Unsandboxed Chromium Process

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/phoenix_scraper.py:48
Finding

Complete Target URLs Are Disclosed to Third-Party Providers by Default

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:58
Finding

Unpinned Third-Party Package Installation Instruction

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (12)

Tainted flow: 'req' from os.environ.get (line 77, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/phoenix_scraper.py (reported line 54)May include surrounding context.

python
req.add_header("Accept", "application/json")
        req.add_header("Accept-Encoding", "gzip")
        req.add_header("X-Subscription-Token", BRAVE_API_KEY)
        with urllib.request.urlopen(req, timeout=TIMEOUT_STANDARD) as r:
            data = json.loads(r.read())
        results = data.get("web", {}).get("results", [])
        if results:

Tainted flow: 'req' from os.environ.get (line 77, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

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

python
req.add_header("Content-Type", "application/json")
        req.add_header("Authorization", f"Bearer {BRIGHT_DATA_API_KEY}")
        timeout = TIMEOUT_RENDER if render_js else TIMEOUT_STANDARD
        with urllib.request.urlopen(req, timeout=timeout) as r:
            html = r.read().decode("utf-8", errors="ignore")
        if html and len(html) > 500:
            method = f"brightdata_{zone}" + ("_render" if render_js else "")

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill documents use of environment variables and multiple outbound network services, but it does not declare any explicit tool scope or allowed tools. That creates an authorization and transparency gap: an agent may invoke network and env-backed behavior without clear policy constraints, increasing the chance of unintended secret access or unsanctioned external requests.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill routes user-supplied URLs and associated request metadata to third-party providers including Brave Search, Bright Data, and potentially X API, but the description does not warn users that this data leaves the local environment. This can expose sensitive targets, tokens in URLs, internal endpoints, or business-intelligence data to external services and logs.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documented failover chain includes Bright Data unlocking and Playwright headless browser execution against JS-rendered or bot-protected sites, but it omits any warning that the skill may perform automated browser actions against defended targets. This lack of disclosure can lead users to trigger legally, operationally, or ethically risky automation without informed consent, and may increase exposure to account/IP blocking or policy violations.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 · references/x-api.md (reported line 14)May include surrounding context.

Lookup user by username

text
GET https://api.twitter.com/2/users/by/username/:username
    ?user.fields=id,name,username,public_metrics

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 · references/x-api.md (reported line 20)May include surrounding context.

Lookup user by username

text
GET https://api.twitter.com/2/users/by/username/:username
    ?user.fields=id,name,username,public_metrics

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 · references/x-api.md (reported line 46)May include surrounding context.

Lookup user by username

text
GET https://api.twitter.com/2/users/by/username/:username
    ?user.fields=id,name,username,public_metrics

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

This code is hard-wired to transmit requested URLs and retrieved content through Bright Data, an external service. That is expected for the feature, but in a reusable agent skill it creates a real data-exposure risk if callers use it on confidential targets, authenticated pages, or internal resources.

Content

Scanner excerpt · scripts/phoenix_scraper.py (reported line 24)May include surrounding context.

python
# ── Config ────────────────────────────────────────────────────────────────────

BRIGHT_DATA_API_KEY = os.environ.get("BRIGHT_DATA_API_KEY", "")
BRIGHT_DATA_ENDPOINT = "https://api.brightdata.com/request"
DEFAULT_ZONE = os.environ.get("BRIGHT_DATA_ZONE", "web_unlocker")
BRAVE_API_KEY = os.environ.get("BRAVE_API_KEY", "")
BRAVE_ENDPOINT = "https://api.search.brave.com/res/v1/web/search"

External Transmission

Medium
Category
Data Exfiltration
Confidence
82% confidence
Finding

The Brave Search integration shares the target site's hostname and full URL-derived search query with an external provider. In this skill's context, that can leak sensitive investigation targets or private resources to a third party, especially when used automatically as the first tier in a failover chain.

Content

Scanner excerpt · scripts/phoenix_scraper.py (reported line 27)May include surrounding context.

python
BRIGHT_DATA_ENDPOINT = "https://api.brightdata.com/request"
DEFAULT_ZONE = os.environ.get("BRIGHT_DATA_ZONE", "web_unlocker")
BRAVE_API_KEY = os.environ.get("BRAVE_API_KEY", "")
BRAVE_ENDPOINT = "https://api.search.brave.com/res/v1/web/search"

TIMEOUT_STANDARD = 30   # seconds
TIMEOUT_RENDER = 60     # seconds (JS render mode)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The scraper sends user-supplied target URLs to third-party services such as Bright Data and Brave as part of its fallback logic, which can disclose sensitive internal URLs, query parameters, or tokens embedded in URLs. In agent-skill context, this is more dangerous because upstream callers may pass private links, intranet targets, or signed URLs without realizing they will be shared externally.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

The module docstring states 'Never returns empty', implying the scraper always yields non-empty content. In practice, both _fail() and the final all-failed return path produce a result dict with html set to an empty string, so the documentation overstates the behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.