Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward Tavily web-search helper; it sends search queries and the configured Tavily API key to Tavily, which is expected for its purpose.
Install only if you are comfortable sending search queries to Tavily and storing a Tavily API key in environment variables or ~/.openclaw/.env. Avoid using it for secrets, private internal URLs, or proprietary text unless that data is allowed to leave your environment.
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
method="POST",
)
with urllib.request.urlopen(req, timeout=30) as resp:
body = resp.read().decode("utf-8", errors="replace")
try:
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
# 支持多 key 轮询：TAVILY_API_KEYS=key1,key2,key3
def _load_keys():
"""从环境变量或 .env 文件加载多 key"""
keys_str = os.environ.get("TAVILY_API_KEYS", "")
if keys_str:
return [k.strip() for k in keys_str.split(",") if k.strip()]
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
if keys_str:
return [k.strip() for k in keys_str.split(",") if k.strip()]
env_path = pathlib.Path.home() / ".openclaw" / ".env"
if env_path.exists():
try:
txt = env_path.read_text(encoding="utf-8", errors="ignore")
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
if keys_str:
return [k.strip() for k in keys_str.split(",") if k.strip()]
env_path = pathlib.Path.home() / ".openclaw" / ".env"
if env_path.exists():
try:
txt = env_path.read_text(encoding="utf-8", errors="ignore")
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
if keys_str:
return [k.strip() for k in keys_str.split(",") if k.strip()]
env_path = pathlib.Path.home() / ".openclaw" / ".env"
if env_path.exists():
try:
txt = env_path.read_text(encoding="utf-8", errors="ignore")
The skill invokes a Python script that uses environment variables, reads a user home directory config file, and performs outbound network requests, yet the manifest declares no explicit tool scope or allowed-tools restrictions. This weakens least-privilege controls and can cause the skill to be granted broader capabilities than users or orchestrators expect, increasing the blast radius if the script is modified or abused.
The description says to use the skill when the user asks to "search the web / look up sources / find links," which are broad natural-language triggers without tighter scope or exclusion conditions. Because no negative examples or narrower activation boundaries are provided, this could cause unintended invocation in loosely related conversations.
The requirements section uses Chinese-only instructional text ("多 key 轮询（推荐）" and "单 key 模式") while the rest of the file is in English, which imposes a language inconsistency without opt-in or documented rationale. This can violate language/locale policy expectations where the skill should not force or assume a specific language unexpectedly.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
import sys
import urllib.request
TAVILY_URL = "https://api.tavily.com/search"
# 支持多 key 轮询：TAVILY_API_KEYS=key1,key2,key3
def _load_keys():
The skill sends user search queries and an API credential to a third-party service without any in-code notice, consent check, or data-sensitivity guard. In an agent setting, users may provide secrets, internal URLs, or proprietary text in a query, and this code will forward that content externally by design.
No suspicious patterns detected.