AI agents call adb_pull_screenshot to retrieve information from LocalAnt without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool reads and retrieves an existing screenshot from an Android device into the workspace. While it is a Read operation (no data is modified or deleted on the source), the severity is medium because screenshots can contain sensitive information (credentials, personal data, application state) that an autonomous agent could exfiltrate.
From the tool's definition Tool name 'adb_pull_screenshot' and description 'Pull the captured screenshot into the workspace directory' indicate it retrieves/extracts a screenshot artifact. The verb 'pull' denotes data retrieval without modification.
Attacks that exploit this kind of access
Pull the captured screenshot into the workspace directory. It is categorised as a Read tool in the LocalAnt MCP Server, which means it retrieves data without modifying state.
Register the LocalAnt MCP server in PolicyLayer and add a rule for adb_pull_screenshot: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches LocalAnt. Nothing to install.
adb_pull_screenshot is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the adb_pull_screenshot rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for adb_pull_screenshot. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
adb_pull_screenshot is provided by the LocalAnt MCP server (yuga-hashimoto/localant). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →