Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed TinyFish API wrapper that sends a user-provided URL and goal to TinyFish and streams the result back.
Install only if you are comfortable sending the target URL, automation goal, and resulting browsing activity to TinyFish under your TinyFish account. Do not use it with secret-bearing URLs, internal systems, regulated data, or confidential prompts unless that external sharing is approved.
The skill invokes shell commands and an external script but does not declare any tool scope or allowed-tools boundary. That omission weakens least-privilege controls, making it easier for an agent or caller to execute the shell-capable skill in environments where the shell access is broader than intended.
The skill instructs users to send a URL and goal to a third-party TinyFish service, but it does not clearly warn that these inputs will leave the local environment. This can cause unintentional disclosure of sensitive URLs, internal targets, credentials embedded in URLs, or confidential task instructions to an external provider.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
# Each SSE frame starts with "data: ". Strip that prefix, drop blank lines and
# keepalive comments, and print the JSON payload one-per-line so callers can
# consume events as they stream in.
curl -sN -X POST "https://agent.tinyfish.ai/v1/automation/run-sse" \
-H "X-API-Key: ${TINYFISH_API_KEY}" \
-H "Content-Type: application/json" \
-H "Accept: text/event-stream" \
This code performs a network POST to an external endpoint and sends potentially sensitive user or system data (url, goal) along with an API credential in the request headers. While the behavior is inferable from the code, there is no user-facing disclosure at the point of execution such as a prompt, visible notice, or warning comment directed at users about transmitting data to a remote service.