Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed remote search/research skill that sends user queries and URLs to AIsa-backed APIs, with no hidden persistence, destructive behavior, or local data harvesting found.
Install only if you are comfortable sending research queries, prompt text, optional system instructions, and submitted URLs through api.aisa.one and downstream providers using your AISA_API_KEY. Do not use it for confidential prompts, internal/private URLs, or authenticated pages unless that data sharing is approved.
The skill declares required environment access and extensively documents outbound network use, but it does not declare an explicit tool scope such as permissions or allowed-tools. In a permissive harness, this can cause the skill to receive broader capabilities than users expect, weakening least-privilege guarantees and making unintended external requests more likely.
The skill markets cross-source search, verification, and relay-based retrieval but does not clearly warn users that their prompts, queries, and possibly referenced URLs will be transmitted to external services. This creates a privacy and data-handling risk because users may supply sensitive research topics, internal URLs, or proprietary text under the assumption the action is local.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \
-H "Authorization: Bearer $AISA_API_KEY"
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST "https://api.aisa.one/apis/v1/perplexity/sonar" \
-H "Authorization: Bearer $AISA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
The extract, crawl, and map examples encourage submitting arbitrary URLs to remote services without warning about privacy, compliance, or impact on third-party systems. This is risky because users may unintentionally transmit internal addresses or sensitive resources, and crawling behavior can increase legal, operational, or reputational exposure.
The skill is designed to send data to https://api.aisa.one, which is expected functionality, but in an agent environment external transmission is still security-relevant. Because the tool accepts arbitrary user queries and URLs and forwards them off-platform, the main risk is unannounced data egress rather than code execution or credential theft.
class SearchClient:
"""AIsa Search API client."""
BASE_URL = "https://api.aisa.one/apis/v1"
def __init__(self, api_key: Optional[str] = None):
self.api_key = api_key or os.environ.get("AISA_API_KEY")
User-supplied queries and URLs are transmitted to external services without any explicit notice, confirmation, or minimization. In a multi-source search skill, users may enter sensitive internal research topics, proprietary URLs, or regulated data, and this design can unintentionally exfiltrate that information to third-party providers.
The code exposes tavily-crawl and tavily-map capabilities that go beyond the stated manifest description of search/retrieval. In an agent-skill context, this scope expansion matters because it enables broader external interaction against arbitrary user-supplied URLs, which can increase privacy, compliance, and unintended network access risk even if the implementation is not overtly malicious.
The top-level usage docstring enumerates supported commands, but it does not list tavily-crawl or tavily-map even though both commands are implemented later and wired into the CLI. This creates intent/documentation drift about what the tool actually supports.
The client reads the AISA_API_KEY from the environment, which is a sensitive credential access path. While the code errors if the variable is absent, it does not include any explicit warning or disclosure that the tool consumes an API key from the environment.
No suspicious patterns detected.