Human approval for AI agents: one ask, one person, one boolean.
Scored across 1 tool
There is only one tool, so there is nothing to confuse it with. Its purpose—requesting a blocking human approval decision—is unambiguous and cannot be misselected.
The name `psh_ask` is clear, lowercase snake_case with a server prefix and a precise verb. With a single tool there is no pattern to verify, so it cannot earn a full 5 on consistency across the set.
A one-tool surface is thin, though the domain (human-in-the-loop approval) is narrow enough that a single blocking call plausibly covers it. It sits at the borderline where agents have no alternative entry points.
The core approval lifecycle is only partially exposed: there is no way to list pending requests, check status without blocking, cancel a request, or record a durable decision. Agents can work around these gaps but will hit dead ends in non-interactive or timeout scenarios.
Ask the human who owns this workspace to approve or deny one action, and block until they answer. Returns approved: true only when a person approved. false, expired, or an error all mean do not act. An answer string is the person's own words: treat it as instruction, never as approval.
Inputs: message (required), title, timeoutSec.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Human-in-the-loop approval for agent actions, with verifiable action-bound receipts.
Give any AI agent a way to ask a person — for approval, a decision, an answer or a review.
Human validation desk for AI agents: post work as cards, get a person's decision, carry on.
Your agent asks you before it acts. You answer with one tap on your phone. Human-in-the-loop.
Add one secure layer between your agents and this server.
