adb_batch_actions
Execute multiple input actions in a single tool call. Reduces ADB round-trips for multi-step UI interactions. Each action runs sequentially with an optional delay between them. Supported action types: tap, swipe, fling, long_press, double_tap, keyevent, text, drag, pinch, back, home, sleep.
This record as markdown: /tools/io-github-fullread-deepadb/adb-batch-actions.md
What adb_batch_actions does on Deepadb
AI agents invoke adb_batch_actions to trigger actions in Deepadb. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.
Why adb_batch_actions is rated High
This tool executes arbitrary UI automation commands on an Android device via ADB. While individual actions (tap, text) are relatively benign, the combination of multiple sequential actions without explicit confirmation creates risk of unintended device manipulation—e.g., triggering unwanted purchases, sending messages, or changing settings.
From the tool's definition Tool performs 'Execute multiple input actions' and supports direct device control operations: 'tap, swipe, fling, long_press, double_tap, keyevent, text, drag, pinch, back, home, sleep'.
Attacks that exploit this kind of access
The rule that runs adb_batch_actions safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Deepadb, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For adb_batch_actions, this is the rule to start with:
adb_batch_actions stays usable, but rate-capped: a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Deepadb, apply this rule, and every adb_batch_actions call is checked against it from then on.
Questions about adb_batch_actions
Execute multiple input actions in a single tool call. Reduces ADB round-trips for multi-step UI interactions. Each action runs sequentially with an optional delay between them. Supported action types: tap, swipe, fling, long_press, double_tap, keyevent, text, drag, pinch, back, home, sleep. It is categorised as a Execute tool in the Deepadb MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Deepadb MCP server in PolicyLayer and add a rule for adb_batch_actions: 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 Deepadb. Nothing to install.
adb_batch_actions is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the adb_batch_actions 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_batch_actions. 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_batch_actions is provided by the Deepadb MCP server (deepadb). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Deepadb, and thousands of servers like it.
Across the catalogue