Replai

Ask the draft Agent

agent_ask

Asks the current draft one ad hoc question, as the dashboard preview does.

Uses the monthly test allowance and records a test result; nothing reaches a customer and simulated actions are not performed. conversation_id from a previous answer continues that preview while the draft is unchanged; history_reset reports a fresh start. A lost response is a new attempt on retry.

Input Schema

Name	Required	Description
`message`	Yes	Ad hoc question for the current draft Agent preview; no customer message is sent.
`agent_id`	Yes	Agent ID from agents_list or agent_create.
`conversation_id`	No	Preview conversation ID from agent_ask; null starts a new preview.

Output Schema

Name	Required	Description	Default
`code`	No
`answer`	No
`status`	Yes
`agent_id`	Yes
`evidence`	No
`recovery`	Yes
`persistence`	No
`history_reset`	No
`conversation_id`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedNov 8, 2026

addedOutput schema / $defs / TestResult / properties / accepted_at

Added value: +{
+  "anyOf": [
+    {
+      "format": "date-time",
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "title": "Accepted At"
+}

Changed3 schema fields changedOct 30, 2026

addedInput schema / properties / agent_id / description

Added value: +"Agent ID from agents_list or agent_create."

addedInput schema / properties / conversation_id / description

Added value: +"Preview conversation ID from agent_ask; null starts a new preview."

addedInput schema / properties / message / description

Added value: +"Ad hoc question for the current draft Agent preview; no customer message is sent."

AddedOct 30, 2026

TDQS

A3.8/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the annotations: discloses that the monthly test allowance is consumed, a test result is recorded, no customer impact occurs, simulated actions are not performed, and that a retry after a lost response is a new attempt (matching non-idempotentHint with concrete consequence). Minor issue: it references 'history_reset', which is not a parameter in the schema.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Four tight sentences, front-loaded with what the tool does, then safety and continuation semantics. Almost no wasted wording.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With annotations and an output schema covering safety and return shape, the description supplies what an agent needs: side effects on quota, no customer exposure, and retry behavior. The unexplained 'history_reset' reference is the only completeness blemish.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already defines all three parameters well. The description adds only the conversation-continuation semantics, which the schema largely repeats; the 'history_reset' mention adds confusion rather than meaning.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource: ask the current draft one ad hoc question, scoped as a dashboard-style preview. An agent can distinguish this from agent_draft_read or test_case_run, though no sibling is named explicitly.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Implies preview/testing usage by noting it consumes the monthly test allowance and never reaches a customer, but it gives no explicit when-to-use versus test_case_run, test_case_evaluate, or conversations_search. The continuation rule via conversation_id is the strongest guidance present.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Replai tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?