PrePublish - YouTube script QA

Audit a YouTube script before recording

audit_script

Run Prepublish's full audit on a YouTube script the user has already written. Returns hook, structure and pacing scores, a script-level attention-risk map naming the passages most likely to lose viewers, and a copy-paste rewrite for each flagged passage. Choose this when the user has a finished or near-finished draft and wants to know what is weak before they record. Do not use it to generate a script, to review a published video, or to check grammar. An email is optional: the result comes back in the conversation either way. This is a text-only check of an unrecorded script. It maps relative attention risk inside the draft. It does not measure or predict published YouTube retention, and it cannot account for delivery, editing, thumbnail, topic or distribution.

Input Schema

TableJSON Schema

Name	Required	Description
`email`	No	Optional. Only when the user gave an address and wants the result link by email: Prepublish stores it with the audit, emails the result link to it, and may send a short series of follow-up emails about the audit that include an unsubscribe link. Leave it out otherwise; the audit runs without it. Never invent an address. When the caller is authenticated with a Prepublish API key the account address is used, so do not ask for one in that case.
`audience`	No	Target audience, only if the user said it.
`category`	No	Video category, only if the user said it.
`script_text`	Yes	The script text as the user wrote it. Paste it verbatim; do not rewrite, summarise or clean it first.
`video_title`	Yes	The title the user plans to publish under. The audit checks whether the script delivers this promise.
`video_duration_seconds`	No	Intended runtime in seconds, only if the user stated one. Used for pacing; never assumed.

Output Schema

TableJSON Schema

Name	Required	Description
`notice`	Yes	The scope limit that accompanies every Prepublish result. Repeat it to the user rather than dropping it.
`status`	No	Present only when the audit had not settled inside this call; the value is "running".
`analysis`	No	The settled audit: status (and error_message when it failed), video_title, overall_score with hook_score, structure_score and pacing_score, ready_to_record, one_key_improvement, score_explanations, the attention_map (position_pct, hold_index, note, sentiment), the improvements in priority order (priority, category, title, location, original_quote and, for the passages on this plan, improved and why), locked_count, and policy_preflight. The script text itself and internal bookkeeping are not included. Absent while the audit is still running.
`next_step`	No	Present only when the audit is still running: what to do to collect it.
`analysis_id`	Yes	Prepublish id for this audit. Pass it to get_audit to re-read the result later.
`open_in_browser`	Yes	Permalink to the same audit on prepublish.ai.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 2, 2026

changedInput schema / properties / email / description

Previous value: -"Where the free result is sent. Prepublish stores this address with the audit, emails the result link to it, and may send a short series of follow-up emails about the audit that include an unsubscribe link. Required for every anonymous audit: the backend rejects the request without it. Ask the user; never invent it. When the caller is authenticated with a Prepublish API key the result comes back in the conversation instead, so do not ask for an address in that case."New value: +"Optional. Only when the user gave an address and wants the result link by email: Prepublish stores it with the audit, emails the result link to it, and may send a short series of follow-up emails about the audit that include an unsubscribe link. Leave it out otherwise; the audit runs without it. Never invent an address. When the caller is authenticated with a Prepublish API key the account address is used, so do not ask for one in that case."

Changed2 schema fields changedSep 27, 2026

changedInput schema / properties / email / description

Previous value: -"Where the free result is sent. Required for every anonymous audit: the backend rejects the request without it. Ask the user; never invent it. When the caller is authenticated with a Prepublish API key the result comes back in the conversation instead, so do not ask for an address in that case."New value: +"Where the free result is sent. Prepublish stores this address with the audit, emails the result link to it, and may send a short series of follow-up emails about the audit that include an unsubscribe link. Required for every anonymous audit: the backend rejects the request without it. Ask the user; never invent it. When the caller is authenticated with a Prepublish API key the result comes back in the conversation instead, so do not ask for an address in that case."

changedOutput schema / properties / analysis / description

Previous value: -"The settled audit as Prepublish returned it: status, overall, hook, structure and pacing scores, the attention-risk map, improvements and, where the plan allows, rewrites. Absent while the audit is still running."New value: +"The settled audit: status (and error_message when it failed), video_title, overall_score with hook_score, structure_score and pacing_score, ready_to_record, one_key_improvement, score_explanations, the attention_map (position_pct, hold_index, note, sentiment), the improvements in priority order (priority, category, title, location, original_quote and, for the passages on this plan, improved and why), locked_count, and policy_preflight. The script text itself and internal bookkeeping are not included. Absent while the audit is still running."

Changed1 schema field changedSep 20, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "properties": {
+    "analysis": {
+      "description": "The settled audit as Prepublish returned it: status, overall, hook, structure and pacing scores, the attention-risk map, improvements and, where the plan allows, rewrites. Absent while the audit is still running.",
+      "type": "object"
+    },
+    "analysis_id": {
+      "description": "Prepublish id for this audit. Pass it to get_audit to re-read the result later.",
+      "type": "string"
+    },
+    "next_step": {
+      "description": "Present only when the audit is still running: what to do to collect it.",
+      "type": "string"
+    },
+    "notice": {
+      "description": "The scope limit that accompanies every Prepublish result. Repeat it to the user rather than dropping it.",
+      "type": "string"
+    },
+    "open_in_browser": {
+      "description": "Permalink to the same audit on prepublish.ai.",
+      "type": "string"
+    },
+    "status": {
+      "description": "Present only when the audit had not settled inside this call; the value is \"running\".",
+      "type": "string"
+    }
+  },
+  "required": [
+    "notice",
+    "analysis_id",
+    "open_in_browser"
+  ],
+  "type": "object"
+}

Changed1 schema field changedSep 20, 2026

changedInput schema / properties / email / description

Previous value: -"Where the free result is sent. Required for every anonymous audit: the backend rejects the request without it. Ask the user; never invent it."New value: +"Where the free result is sent. Required for every anonymous audit: the backend rejects the request without it. Ask the user; never invent it. When the caller is authenticated with a Prepublish API key the result comes back in the conversation instead, so do not ask for an address in that case."

Changed1 schema field changedSep 19, 2026

changedInput schema / properties / email / description

Previous value: -"Where the free result is sent. Required for every anonymous audit — the backend rejects the request without it. Ask the user; never invent it."New value: +"Where the free result is sent. Required for every anonymous audit: the backend rejects the request without it. Ask the user; never invent it."

First observedSep 10, 2026

TDQS

A4.6/5.0

Behavior5/5

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

Annotations only say readOnlyHint=false, openWorldHint=true, destructiveHint=false; the description adds substantial non-obvious behavior: the email address is stored with the audit, a result link is emailed, follow-up emails with an unsubscribe link may be sent, and the account address is used under API-key auth. It also scopes limits precisely (text-only, unrecorded script, relative in-draft attention risk, no retention prediction, no delivery/thumbnail/topic factors).

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

Conciseness4/5

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

Front-loaded with purpose and outputs, then usage rules, then caveats. Efficient overall, though the limitation block is slightly repetitive ('text-only check of an unrecorded script' overlaps with the earlier draft framing) and costs more lines than strictly necessary.

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

Completeness5/5

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

For a multi-field audit tool with an output schema, the description covers what it returns, when to invoke it, what it will not do, and the email side effect, and it explicitly bounds the meaning of the attention-risk score. Nothing needed to invoke it correctly is missing.

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

Parameters4/5

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

Schema description coverage is 100%, so the schema already carries parameter meaning (baseline 3). The description adds value on the email parameter's side effect on delivery (the result returns in the conversation with or without it) and reinforces the 'only if the user said it' convention for audience/category/duration.

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

Purpose5/5

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

States a specific verb+resource ('run Prepublish's full audit on a YouTube script') and enumerates the concrete outputs: hook/structure/pacing scores, an attention-risk map naming passages, and per-passage rewrites. Clearly distinguishable from siblings like audit_hook, script_runtime, or check_authenticity.

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

Usage Guidelines4/5

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

Gives an explicit use condition (finished or near-finished draft, wants to know what is weak before recording) and explicit exclusions (not for generating a script, not for reviewing a published video, not grammar). However, it does not route the excluded cases to the sibling tools that handle them, so it stops short of full when/when-not/alternatives guidance.

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

All PrePublish - YouTube script QA 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?