ClawHub
Back to skill

Security audit

youtube-search-api-skill

Security checks for vulnerabilities and agentic risk

Overview

The skill does the YouTube search extraction it claims, but it encourages users to provide a BrowserAct API key directly to the agent.

Review before installing. Use this only if you are comfortable sending YouTube search terms and a BrowserAct bearer token to BrowserAct. Configure BROWSERACT_API_KEY through a protected environment or secret manager, do not paste it into chat, and stop the process manually if polling runs too long.

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 (2)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:27
Finding

API Key Disclosure Through Agent Conversation

Content
View full analysis
"Since you have not configured the BrowserAct API Key, please go to the [BrowserAct Console](https://www.browseract.com/reception/integrations) first to get your Key." ``` `scripts/youtube_search_api.py:86-90`: ```python if not api_key: print("\n[!] ERROR: BrowserAct API Key is missing.", flush=True) print("Please follow these steps:", flush=True) print("1. Go to: https://www.browseract.com/reception/integrations", flush=True) print("2. Copy your API Key.", flush=True) print("3. Provide it to me or set it as an environment variable (BROWSERACT_API_KEY).", flush=True) ``` ### Technical Analysis The Skill directs the Agent to request the user's collaboration in providing an API key, while the script explicitly tells the user to “Provide it to me.” This can encourage the user to paste a long-lived BrowserAct credential into an Agent conversation. A credential entered into a conversation may be retained in chat history, model traces, platform telemetry, debugging records, or other logs. This exposure is unnecessary because the implementation already supports reading the key from the `BROWSERACT_API_KEY` environment variable. The script does not print the key after reading it, and its transmission as an HTTPS Bearer token to the declared BrowserAct API is necessary for the stated functionality. The vulnerability is therefore in the credential-provisioning guidance, not in the authenticated API request itself. ### Attac ...[truncated 1099 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Note
Location
scripts/youtube_search_api.py:28
Finding

Unbounded HTTP Requests and Task Polling

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill requires environment access to a secret (BROWSERACT_API_KEY) and performs network-backed data extraction, but it does not declare any explicit tool scope or allowed-tools boundary. That omission increases the chance an agent can invoke broader execution or network capabilities than intended, weakening least-privilege controls and making misuse or accidental overreach harder to contain.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill description repeatedly instructs the agent to 'proactively apply' the skill across a wide range of ordinary research tasks, which can cause over-triggering. In practice, this may lead the agent to send user queries and derived research activity to an external service without sufficiently specific user intent or necessity, increasing privacy, cost, and data-governance risk.

Content

No source excerpt is available for this finding.

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/youtube_search_api.py (reported line 15)May include surrounding context.

python
# API Configuration
TEMPLATE_ID = "82165150735472749"
API_BASE_URL = "https://api.browseract.com/v2/workflow"

def run_youtube_search_task(api_key, keywords, video_type="Videos", limit=100):
    headers = {"Authorization": f"Bearer {api_key}"}

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/youtube_search_api.py (reported line 31)May include surrounding context.

python
# 1. Start Task
    print(f"Start Task", flush=True)
    try:
        res = requests.post(f"{API_BASE_URL}/run-task-by-template", json=payload, headers=headers).json()
    except Exception as e:
        print(f"Error: Connection to API failed - {e}", flush=True)
        return None

Ssd 3

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script explicitly tells users to provide the BrowserAct API key directly to the agent, which encourages secret disclosure through a conversational channel rather than a secure secret store. In an agent setting, this creates a real risk that credentials are exposed in chat logs, telemetry, prompts, or downstream tool calls, enabling unauthorized use of the external account.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.